16th International Symposium on "Yacht Design and Yacht Construction"
Heer, P.W. de, Editor
Report 1239-P November 2000
TU Deift Faculty of Design, Engineering and Productiony Department of Marine Technology Deift University of Technology Ship 1-lydronsechanics Laboratory 16th International Symposiumon "Yacht Design and Yacht Construction"
Amsterdam, 13 November 2000
PROCEEDINGS
Edited by P.W. de Heer
October 2000
Organized by HIS WA - National Association of Watersport in The Netherlands, the International Trade Show for Marine Equipment METS 2000 and the Deift University of Technology
Deift University of Technology Ship Hydromechanics Laboratory Printed by:
DocVision BV Leeghwaterstraat 42 2628 CA Deift The Netherlands
Telefoon: +31 15 2784642 Fax: +31 15 2781749
CIP-DATA KONINKLIJKE BIBLIOTHEEK, DEN HAAG
16th International Symposium on "Yacht Design and Yacht Construction": proceedings of the 16th International Symposium on "Yacht Design and Yacht Construction", Amsterdam 13 November 2000/P.W. de Fleer(editor),-DeiftUniversityof Technology,Ship Hydromechanics Laboratory, The Netherlands. ISBN: 90 - 370 - 0185 - 8 Subject headings: Yacht Design, Yacht Construction TABLE OF CONTENTS
PROGRAMME 5
INTRODUCTION 7
THE VERIFICATION OF MAST AND RIGGING OF LARGE SAILING VESSELS Michael J. Gudmunsen, Lloyd's Register, London, England 9
PRACTICAL EXPERIENCE ON REDUCING MOTIONS AND IMPROVING COMFORT ON BOARD LARGE MOTOR YACHTS H.M. van Wieringen, F.A. Gumbs, F. De Voogt, International Ship Design and Engineering, Bloemendaal, The Netherlands R. Dallinga, MARiN, Wagenin gen, The Netherlands 51
SOME CRITICAL NOTES ON DESIGNING WITH COMPOSITES Jons Degrieck, Ghent University, Ghent, Belgium 65
ALL ELECTRIC YACHT- ELECTRIFYING OR TERRIFYING? U. Nienhuis, Netherlands Institute for Maritime Research, The Hague, The Netherlands 77
PERFORMANCE PREDICTION OF SCHOONERS USING WINDTUNNEL DATA IN VPP CALCULATIONS I.M. C. C'ampbell. Wofson Uiit MTL4, Southampton, U'?ited Kingdom G. Dijkstra, Gerard Djkstra and Partners, Amsterdam, The Netherlands 91
THE INFLUENCE OF BOW SHAPE ON THE PERFORMANCE OF SAILING YACHTS J.A. Keuning, R. Onnink, A. Damman, Ship Hydromechanics Laboratory, DeIft University of Technology, Deift, The Netherlands 107 PROGRAMN'IE
16th International HISWA Symposium on "Yacht Design and Yacht Construction".
Monday, 13 November 2000
JADE LOUNGE
08:00 - 10:00 Registration and information
ROOM RIS
10:00 - 10:15 Moderator Jack A. Somer
Word of welcome by Jack A. Somer
10:15 - 11.00 Michael J. Guthnunsen, Lloyd's Register, London, England
THE VERJFICA T1ON OF MAST AND RIGGING OF LARGE SAILING VESSELS
11.00-11.30 Break
11:30 - 12:15 H.M. van Wieringen, F.A. Gumbs, F. De Voogt, International Ship Design and Engineering, Bloemendaal, The Netherlands R. Dallinga, MARIN, Wageningen, The Netherlands
PRA CTI CAL EXPERIENCE ON RED UCING MOTIONS AND iMPROVING COMFORT ON BOARD LARGE MOTOR YACHTS
12:15 - 13:00 Jons Degrieck, Ghent University, Ghent, Belgium
SOME CRITICAL NOTES ON DESIGNING WiTH COMPOSITES
13.00 - 14.00 Lunch in Parkrestaurant?
5 14.00 - 14.45 U. Nienhuis, Netherlands Institute for Maritime Research,The Hague, The Netherlands
ALL ELECTRIC YACHT- ELECTRIFYING OR ThRRIFYING?
14.45 - 15.30 I.M.C. Campbell, Wolfson Unit MTIA, Southampton, United Kingdom G. Dijkstra, Gerard Dijkstra and Partners, Amsterdam, The Netherlands
PERFORMANCE PREDICTION OF SCHOONERS USING WIND TUNNEL DATA IN VPP CALCULATIONS
15.30 - 16.00 Break
16.00 - 16.45 J.A. Keuning, R. Onnink, A. Damman, Ship Hydromechanics Laboratory, Deift University of Technology, Delft, The Netherlands
THE INFLUENCE OF BOW SHAPE ON THE PERFORMANCE OF SAILING YACHTS
16.45 - 17.00 Closing
JADE LOUNGE
17.00 - 18.00 Reception
6 INTRODUCTION
On behalf of the Organizing Committee ot the 16th International HISWA Symposium on Yacht Design and Yacht Construction I have the pleasure of invitingyou to participate in the forthcoming Symposium.
The Organizing Committee believes that it succeeded in getting togetheran interesting set of papers presented by well-known experts in their fields. This year, the design and construction of large custom-built yachts is emphasised. The construction of largecustom-built yachts, both motor yachts and sailing yachts, is an ever-increasing market, in particularover the last decades, and an area in which high-tech developments playan important role both in realising these projects as well as in acquiring them.
This year, the set-up of the Symposium is slightly changed to allow formore time in-between the sessions, so that ample time is available for informal contacts between the delegatesand/or the presenters of the various papers. Since yacht designers and researchers from allover the world attend the Symposium it never fails to be an interesting day for all ofyou. We hope to have the pleasure of welcoming you at the 16th Symposium.
Alexander Keuning
7 THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
By M. J. Gudmunsen Principal Surveyor Lloyd's Register Marine Division
9
THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
The Verification of Masts and Rigging of Large Sailing Vessels by M. J. Gudniunsen Senior Surveyor Lloyd's Register Marine Division
SYNOPSIS
This paper presents the formal classification approach used in the appraisal and verification of the masts and standing rigging of large sailing passenger vessels. Comparison is made between the tabular scanthngs for masts and rigging in Lloyds Register's rules and regulations of 1922, and the direct calculation techniques currently employed by Lloyds Register.
The calculation methods and assumptions are described with illustrations from analyses of the rigs of both newbuilding vessels and vessels currently in service.
Amongst the number of large sailing vessels rigs referenced, examples from the "STAD AMSTERDAM" rig have been presented, as it represents the visual replication of a traditional 19th century sailing ship but employs a range of modern 20th century hi-tech materials.
The criteria for rig design, verification and acceptance into class is presented together with details on the build quality, materials, survey, testing and certification requirements.
Designers, tasked with meeting dassification or flag administration requirements for large sailing vessels have foand the criteria presentedinthis paper useful with regard to setting a standard for both design and acceptance of the masts and standing rigging.
(t-- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
INTRODUCFION At this time, the Langebrugge shipyard in Belgium had received an During the second half of the 1980's order from White Star Clippers of interest was developing with owners Brussels to build two barquentines, and operators in passenger ships suitable for up to 194 passengers with propelled by wind power. The a crew of 59. The ships were to concept was to provide a high measure 111.57m over the bowsprit, standard of accommodation and have a loaded displacement of 2556 services combined with the unique tonnes and carry 3365 m2 of sail area experience of tall ship sailing for on four steel masts. normal fare paying passengers. Initially intended for operation in the Caribbean the first ships would operate out of Miami on a pattern of one or two-week cruises. Research conducted by White Star suggested that their guests would be predominantly European, with a knowledge and interest in sailing The rig would require to be designed in order to permit some participation on the part of the passengers but with a high degree of confidence in the safety figure 1 of the rigging and sail systems.
The first vessel, "Star Dipper" leaving Flushing for her sea trials is shown in figure 1. The vessels were submitted for formal classification with Lloyds Register with a contemplated hull notation of 100A1 Sailing Passenger Ship.
Verification of and acceptance into class of the all-welded steel hull of the vessels was facilitated through the application of the Rules and Regulations for the classification of Steel Ships with some minor amendments.
The verification of the masts and rigging, however, presented a more complex structural problem. Lloyd's Register's Rules for masts and standing rigging of sailing vessels had been discontinued in the mid 1920's. These Rules had provided scantlings and dimensions of masts, yards and wire rope standing rigging in tabular form. Mast scantlings being based upon mast length and rigging dimensions being based upon a numeral using the ship's principal dimensions.
These early rules were developed and maintained mainly through service experience. Théicantlings of masts and wire ropes were based upon the material performance available at that time and the construction methods employed such practices as riveting of curved shells to form tubular mast structures. From the 2- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS tabulated scantling method, it is clearly not possible to determine the margins of safety inherent in the final design, nor establish the limiting environmental conditions for the rig.
The tabular scantling approach provides no flexibility regarding design change or novel design features and would appear to be adequate solely for a traditional mast/rigging configuration.
Interestingly, in parallel with the appraisal of the Langebrugge barquentmes, another large LR classed sailing passenger ship, "Le Ponant" was under development at the French shipyard of SCNF.
This aluminium hulled vessel was proposed with high strength oval section aluminium grooved masts, stainless steel rod rigging1 hydraulic mast jacking and hydraulic sail systems. The arrangements would permit most sailing operations to be conducted solely by the helmsman from the central steering position.
The design employed over 2000m2 of sail area, fore and aft rigged on the 3 masts. figure 2
Clearly, from the diverse nature of the mast and rig designs being contemplated, a more direct calculation approach was demanded in order to verify the structural adequacy of the arrangements and to provide a satisfactory procedure for formal classification.
The mast and rigging scanthng tables from Section 37 of the 1922 Rules and reproduced in Appendix 1. In order to LiLILaLt LULLLj)CU1J1L YVIUL ilLU.Lt11L"b t.ULUuìb, utLaL)1c) 1Lac LCtflau..uuunaky produced in metric format and are annotated with the table reference number and 'Imperial' or 'Metric' as appropriate.
Many terms stated in the rules and in the tables may be unfamiliar to today's Naval Architects, namely, mast partners, hounds, futtocks, chainpiates, topgallants, royals, shrouds etc.It is however anticipated that the reader will consult the many THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS reference books available on this topic in order to develop familiarity with these traditional nautical terms. TUE VERIFICATION OF MASTS 4ND RIGGING OF LARGE SAILING VESSELS
ACCEPTANCE CRITERIA
From a Classification Society perspective, the structural verification of the masts and rigging is carried out in order to establish the level of safety and reliability of the structures and the supporting rigging components. Many large sailing ships are used for training purposes as well as those dedicated to providing a "tall ships" experience for physically disabled people. At the other end of the client spectrum, we have those sailing ships that are dedicated passenger ships, carrying significant numbers of people who range in age from children to the very senior age groups.
For all these vessels, failure of the rig, masts or supporting structures could have far reaching consequences. Recently built sailing ships generally have an auxiliary engine which is used to power the vessel in rivers, harbours and estuaries, or in wind conditions where the sails cannot often be utilised. When utilising wind power, the sail system must be considered as the "engine" of such ships and loss of this prime mover may endanger the survivability of the complete vessel. Failure of individual rigging components or masts would endanger the lives of crew and passengers and it is not possible to provide any real physical separation between the personnel and the location of the risk.
The proximity of booms, blocks and other rigging to passengers and crew is clearly illustrated in figure 3 which shows Star Clipper underway.
As previously mentioned, the tabular methods of mast and rigging scantling selection provide the designer with no indication of the inherent factors of safety against failure.
Where rigs are proposed with modified, hybrid or indeed novel arrangements, empirical derivation of scantlings would be highly unreliable.
figure 3 In order to establish suitable factors of safety for rigging, standards used in the marine industry for lifting appliances were examined. Stayed derrick posts are similar in configuration and response to stayed masts and it was considered that the THE VERiFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS accepted safety criteria for sailing vessel masts and rigging should, at the very least, comply with these criteria.
Since the prime purpose of the standing rigging is to provide support to the mast and bowsprit structures, it would be reasonable to ensure that the standing rigging had a moderate to high factor of safety against failure. With this rationale, stresses in the mast structures could be allowed up to a level normally accepted by classification for integral components of the ship's structure. Any selected factor of safety must be commensurate with the probability of design load exceedence and hence the following safety factors against failure of the standing rigging were established;
All sailing conditions F.O.S=3.5 Survival, bore poles conditionF.O.S= 2.0
The masts and bowsprit are subject to a combination of both axial compressive and bending loads. Factors of safety must therefore be related to the critical axial failure load (stress) and also the bending failure load (stress). Hence, the mast acceptance criteria are based upon a combined safety index:
All sailing conditions [ab!a+aa/ a] 0.67
Survival, bare poles condition [ab /a+ aal (Ycrit] 0.85
Where :- abis the derived bending stress in the mast. aais the derived axial stress in the mast. a is the material yield stress. aaitis the critical elastic buckling stress.
I-- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
MATHEMATICAL IDEALISATION
With the complex interaction between mast and rigging stiffness and themteraction between successive masts due to their connection via longitudinal stays, it is considered that any simplistic analytical calculation would be both maccurate and unsuitable.
The verification process requires calculation of the component tensile loads for the rigging and the axial force and bending moment results forevery section of the masts and bowsprit. The analysis technique employs a large displacement non- linear finite element code which satisfies both the response and output result requirements.
The mast and bowsprit structures essentially perform as axially loaded compression columns. Due to the load contribution of the running and standing rigging at various points on the masts, each different section of mast between rigging attachment points has its own unique set of axial and bending loads. Hence, the masts and bowsprit are represented mathematically by line beam elements, each with its commensurate axial, bending, shear and torsional geometric properties.
The standing rigging is represented by one-dimensional line elements attached to the masts and to the deck or chainpiates as defined on the rigging plan. Since rigging is unable to provide compressive stiffness, the associated material is provided with a non-linear response capability, representing the normal load- extension response in the tensile domain but a zero load-extension response in the compression domain. The axial stiffness of the rigging components depends upon the basic Young's Modulus of the material of construction, the effective cross section and the efficiency of the geometric section to resist strain. With steel wire rope, around 20% of the axial stiffness is lost due to the lay of the rope and further losses are incurred due to the difference between net cross section and gross idealised cross section.
In consideration of these effects, the following list indicates acceptable Young's Modulus values for a range of rigging materials based upon nominal diameter for the derivation of cross sectional area:
Component material Young's modulus Area based upon (N/mm2) (mm2) MP 35N 2.32E5 Nominal dia Stainless steel (22-13-5) 1.92E5 Nominal dia Titanium (6AL-4V) 1.10E5 Nominal dia Aluminium (6061-T6) 0.72E5 Nominal dia G.S.W.R 1.09E5 Nominal dia S.S.W.R 1.21E5 Nominal dia Alpha & Gamma rods 1.93E5 Nominal dia Chain 2.06E5 Nominal dia THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
A 3-Dimensional model of the masts and rigging for the Stad Amsterdam is shown in figure 4. The lower masts are of steel, whereas the topmasts and topgallant masts are of high strength aluminium ahoy. The steel bowsprit has additionally a jib- boom, again fabricated from high strength aluminium alloy. The analysis model shown has 189 nodes, 92 axial/bending mast elements and 154 axial rigging elements.
figure 4
t THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
LOADING DETERMiNATION
Other than motion induced loads, the significant loading on the masts and rigging results directly from the wind speed and direction. In order to take dueaccount of the relative velocity between the ship and the wind, the sail loadsare based upon an apparent wind velocity (Va). Clearly, there is no effective means of controlling Va and adjustment to sail area is the logical reaction to increasing values of apparent wind speed. In very severe wind and sea conditions the masts and rigging must be capable of survival without sustaining excessive loading. Hence, a "bare poles" condition with a hurricane wind speed of 122 knots from any direction is considered as a survival load case. With only "bare poles and furled sails, the vessel will also be subject to wave induced ship motions, particularly roll which will induce lateral acceleration forces into the masts. Three clearly defined limiting conditions result from the anticipated operational modes of large sailing vessels;
Normal operation with a full press of sails.
Storm conditions with reduced sail.
Survival condition with all sails furled.
Depending upon the rig configuration, fore and aft sails or square sails, the design will have a stated apparent wind speed in association with sailing condition 1). Usually this is defined by the Beaufort wind scale and for classification purposes is not taken as less than 25 knots,
The storm condition wind speed is largely dictated by the size and number of sails which can be effectively used in high wind conditions. Of the vessels analysed to date, this storm wind speed is generally between 40 and 50 knots. Again, for classification purposes the storm apparent wind speed is not taken as less than 40 knots. In addition to the wind generated loads in the survival condition, long term values of ship motion are used as follows:-
Ship Motions Motion I\Iaxinìum Single Period ìn Amplitude Seconds k(oU Ø=3O Tr = 0.7B/(GM)0.3
Pitch = l2exp (Lbp/ 300)T = 0.5 (Lbp)°.S
Heave Lbp/80 Th = 0.5 (Lbp)0.5
t3 THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Where: Lbp = Length of ship between perpendiculars, in metres, B = moulded breadth of ship, in metres, GM = transverse metacentric height of loaded ship, in metres = is to be taken as not greater than 80.
Wind generated loads carried by the masts from the sails or to staysail stays may be evaluated using typical expressions for the lift on aerofoil sections:
F5 = P/2.AS.VA2.0
Where: = p the density of air. = A5 sail area. = VA Apparent wind speed. CL or CD as appropriate.
The mean lift coefficient (CL) from sails assuming the optimum angle of attack are generally taken to be: 1.1 for staysails 1.2 for fore and aft sails on masts with booms 1.4 for square sails
The drag coefficient for masts, standing rigging and yards with furled sails is taken to be 1.2. Application of drag is generally only required for the survival condition with bare poles and a very high wind speed.
Staysail loads induce catenary response into the stays and equilibrium between the stay sag, applied sail luff load and stay tension has to be maintained. There clearly is a practical limit as to how much sag can be removed from a staysail stay before abnormally high stay tensions are induced. Over the years there has been much argument and debate over a suitable value of stay sag and it is suggested that the likely value lies somewhere between 3% and 6% of the stay length. For the purposes of calculation, an assumed stay sag within these boundaries is used to develop staysail stay tensions. Hence, staysail stay tension is approximated to:
T9= 3.75 X F
The vectorial components of the stay tensions are added to the loading in the analysis model at the mast and bowsprit attachment points. For analysis of the loads in the staysail stays, the tension induced by the staysail however, must be added to the F.E. analysis result in order to obtain the total stay tension. Although running rigging is not included in the classification process, significant additional loads are generated on the masts and bowsprit at running rigging attachment points. The mass of the yards and sails also contribute to the axial
2o THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS loading in the masts and more importantly, to the lateral forces appliedto the masts as a result of ship motion induced acceleration.
The detailed methods of load application to the finite element modelare not presented in this paper as it is considered that experienced analysts would apply suitable techniques which would generally be discussed and agreed with the certifying authority. THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
MSC'PATRAN Vercion 9.0 28Jun-86 15:25:38
Vector: SC1:2BKNOTS RUNNING. A1:Nonlinear- 188. ' of Load. Applied Loads, Translational, a
226851
218647
1 94444
170240
162036
145033
129629
I 13425
97222
81810
64814
401311
32407 I 16203
O default_Vector Ma, 243055 Nd 614 O Nd 106
figure 6
22 THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
ANALYSIS
The rig analysis forms the basis for ail subsequent appraisal of the masts and standing rigging. The results provide evidence that the acceptance criteria have been satisfied and permit detailed strength assessments to be carried out for complex structures and attachments. The deformed shape of the rig is shown in Figure 7 and represents the loading case of wind from abaft at 28 knots (relative).
NSC/PflTRHN Vcron 9.6 26-Jun-66 15:26:22 DeÇorn,: SCI:28KNOTS PUNNING, R1:Non-11r,r: 166. of Lo8d, T-s1at1ona1, CNONI-LRYI
defau1t_Uefornaton t1 3.19+62 Nd 617
figure 7
The analysis is carried out without the effects of pre-tension in order to consider the likely scenano of a rig in service prior to a scheduled maintenance period. With square rigged ships, the vector of the applied loads is along the ships fore and aft axis, which produces deformations in the x-z plane. With longitudinally rigged vessels such as schooners and barquentines, significant deflection of the masts occurs additionally at 90° to the wind direction. The deformation characteristics are controlled by the relationship between mast and rigging stiffeness. For Stad Amsterdam the topmasts and topgallant masts are of aluminium and due to its low Young's Modulus, relative to steel, the topgallant
-22- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS backstays provide high restraining forces which effectively pull the mast heads aft and result in forward curvature of the mast. The deflected shape of the bowsprit and jib-boom are reproduced in figure 8, for a condition where all the forward staysails are employed. The upward bending of the structure due to the significant stay tensions is apparent. The need for high stiffness in the stays of the bowsprit and jib-boom is clear from this plot in order to reduce bending in the structure. Pre-tensioning of the bobstay and martingales in order to induce downward bending of the bowsprit and jib-boom may be applied in order to attempt to reduce the bending response when under load.
9.0 20 D. SC 2 OTPL*4ING. fln..2: 1 2 o2 L.d. IpI. nt.fr I.tton.I. (NON-t.RYI
£1.,.3.29.02 SNd SI?
figure 8
Typical mainmast scantlings are shown in figure 9. For comparison, the scantling requirements of the 1922 Rules have been indicated and for the steel masts the comparison is favourable. For the topmasts and topgallant masts however, the dimensional comparison between the steel requirements and the alununium scantlings fitted is invalid.
The axial fuie. in the masts ùbtairied for a typical 2 knot wind speed sailing condition for the sailing ship 'Stad Amsterdam' are shown in Figure 10.
In addition to the verification of the mast scantlings and its ability to efficiently sustain this column load, the supporting ship structure requires to be examined in way of the mast heel support and also in way of the attachments for rigging.
2L THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Based upon the calculation of the safety indices described in the section on acceptance criteria, the values obtained for the masts of the example vessel, Stad Amsterdam are reproduced in figure 10, for a typical 28 knot wind speed sailing condition.
290x8 Alu. Fitted 203x5 (1922 Rules) Axial force 13.Om QL11cM Safety Index 0.34
360x10 Alu fitted 229x5 (1922 Rules)
416x12 Alu fitted 533x8 (1922 Rules)
Axial force 11.3 m 711 kN. Safety Index 0.33 (FOS 2.O
420x12 Alu fitted .5?,4x9(l922Rijlps'
569x10 fitted li 576x9 (1922
Axial force Safety Index 909 kN. 13.0 0.48 (FOS1.4)
660x9.5 fitted 694x11 (1922
figure 9 figure 10 Paee 11 -25- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
MAST CONSTRUCTION
Based upon the most severe axial and bending loads obtained from the finite element analysis, the mast sections are verified for compliance with the safety indices set out in the section on acceptance criteria. Generally, tubular masts of circular cross section should have wall thicknesses of greater than D/80. This will ensure that local instability of the mast wall is not a likely premature failure mode.
The shells of circular section masts are generally rolled from plate and one longitudinal seam weld is made to complete each mast section. Successive mast sections should have seams displaced circuferentially by 120°.
By comparison with 19th century construction methods, typically the mast sections would have been formed from three curved plates, each completing a 120° arc, overlapped and double riveted. Successive mast sections were likewise double riveted butin way of the partners and other critical sections, triple riveting was applied. Internal stiffening was provided by angle bars with their flanges riveted through the mast walls. Figure 11 shows a mast figure 11 constructed in this way.
Modern masts are generally of the single pole type, tapered over their entire length. The taper is achieved by either manufacturing mast sections of conical form or connecting constant diameter sections via conical reducers as is the method used for the masts of Royal Clipper.
At the mast heel, suitable arrangements are required to prevent rotation around the mast's own axis and diaphragms or other stiffening are required to stabilise the mast wall in way of the reaction at the mast heel. In view of the high mast heel reaction force, almost all masts are carried through the decks and are supported at the ship's keel. - As the mast sections between decks are short in length and have small end bending moments there is good technical justification for a reduced geometric section in these locations. On the Star Clipper/Star Flyer a reverse conical taper was arranged THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS below the weather deck and a pifiar from the lower deck to the keel. This mininilsed the width requirements for the central passageway in way of the masts. At deck levels where lateral support is provided, the structural arrangements must be such as to prevent localised loads on the mast wall which could lead to deformation of the section. Where the mast is not jointed above the deck a substantial deck insert with spigot is arranged in way of an increased mast wall section to provide an annulus for wooden or plastic wedges. In order to avoid this feature, Royal Clipper has a flanged and bolted connection immediately above the weather deck as indicated in figure 12. Hence the lower section of mast can be fully welded into the hull. Un-stepping the masts for inspection or repair is made easier by this flanged arrangement. figure 12
On traditional tall ships, the standing rigging was passed around the masts and small pegs or "thumb cleats" were arranged to locate the upper ioop of the wire stay or shroud.
With this arrangement, the shroud/stay loading applied to the mast results in a circumferential compressive hoop stress in the mast wall. This principle is visible in figure 13.
figure 13
Most modem sailing vessels employ eyeplates welded to the mast wall in order to allow direct connection of the standing rigging rope terminal or shackle. The capability of the eyeplate should be in excess of the breaking strength of the rope and requires to be of sufficient thickness to provide adequate bearing length for the terminal/shackle pin and to ensure that the pin is not subject to significant bending.
27 THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
The angularity of the shrouds/ stays and the geometry of the eyeplate will result in punching shear loads in the mast walls at the extremities of the eyeplates.
Therefore, in way of shroud and stay attachment points it is required that the mast should be protected by one of the following options:
Internal diaphragms at the upper and lower boundaries of the eyeplates.
External mast rings in way of the upper and lower boundaries of the eyeplates. (see figure 14).
A mast section of 2 x the basic mast wall thickness over a length of 3 x depth of eyeplates.
In all cases, full or partial penetration welds with appropriate preparations are required in order to achieve sufficient shear area for eyeplate attachments. The choice of the most appropriate option depends upon many factors, including mast diameter and aesthetic considerations. The moulded dimension of masts is conventionally the internal surface and hence any steps in mast wall thickness will be external. Where the thickness differential between successive sections exceeds 4mm the figure 14 thicker part is to be tapered with a 1:3 ratio in order to avoid abrupt discontinuity in the section.
The attachment of isolated eyeplates using, doubling plates is permitted provided the eyeplates are first welded to the mast wall and the doubler is slotted over the eyeplate and fully welded to both eyeplate and mast. From a purely engineering point of view, the application of higher strength steels for masts appears initially not to provide significant structural advantage since the elastic buckling response of mild and higher tensile steels are similar. However, since modern all welded mast structures have significant bending stiffness, the chosen mast material must have adequate tolerance to bending induced direct stress. For steel masts, material with a yield stress in excess of 280N/mm2 is frequently selected, with higher tensile steel of 355N/ mm2 yield strength being employed in several designs.
-29- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
STANDING RIGGING
The majority of sailing ships dealt with recently, employ galvanised steel wire rope for the standing rigging although stainless steel is being increasingly used in view of its good corrosion resistance.
As there is usually no requirement for standing rigging wires to pass through blocks or sheaves, the make up of the rope section often consists of small numbers of individual wires.
A typical lower shroud arrangement during assembly is indicated in figure 15.
figure 15
The Jubilee Sailing Trust vessel, "TENACIOUS" uses a316S31 fully austenitic stainless steel wire of i x 19 construction for allstanding parts of the rig. On occasions where running backstays or forestays arearranged, the selected rope is required to provide a degree of flexibility and both"STAD AMSTERDAM" and "CISNE BRANCO" utilise a 6 x 36 construction inthese locations and for the majority of the remaining standing rigging.
It is widely accepted that fibre cored rope is notsuitable for standing rigging and hence is not permitted by many classificationauthorities and flag administrations. Also, some aspects of the material strength usedin the construction of the rope wire appear to be of concern to someadministrations and builders. Although stainless steel wire rope offers acceptable strengthcharacteristics coupled with generally good corrosion resistance, some grades are prone tofatigue cracking caused by stress corrosion. The onset of failureof stainless steel wire ropes is often not readily observed and this has been clearly demonstrated byinstantaneous failure during the break tests of completed ropes and terminal endconnections.
Galvartised steel wire rope (GSWR) is available with wirestrands of three material strengths, 1420N/mm2, 1570N/mm2 and 1770N/mm2.Owner or designer preference dictates which wire material strengthshould be used. As the value of THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Young's Modulus is the same for each strength, larger diameter, low strength wires will provide greater axial stiffness and will in turn provide greater restraintto mast deflection. The non-availability of higher strength GSWR in the specificareas of the world where the vessel is intended to operate may also limit the choice of wire material strength. The use of low strength wire material shouldensure that the fatigue failure of terminal end connections and of loose gear is reduced due to the lower operational stress levels.
The bowsprit and jib-boom, where fitted, are subject to significant loading. The fore royal, topgallant topmast and jibstays are attached to these structures or pass over built in sheaves. These stays are heavily tensioned due to the catenary loading induced by the staysails. The mechanical advantage obtained by the angularity of the bowsprit supporting stays is usually not optimal and results in high tensions in the bobstay and figure 16 martingale stays.
Invariably, these are often of solid round bar section of significant diameter or short link steel chain. Figure 17 shows a typical rod arrangement and figure 16 a short link chain arrangement. Figure 18 shows the stem connection of a wire rope bobstay.
figure 17
-
figure 18
30- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
TESTING, SURVEY AND CERTiFICATION
Any formal design appraisal is required to be complimented by a suitable regime of testing and survey in order to certify the structural components of the rig as 'fit for purpose.'
The masts and bowsprit may be fabricated from tubular steel or aluminium. The mast sections may employ classic yacht oval aluminium extrusions or alternatively may be more complex space frames or assemblies employing hi-tech materials.
For the large passenger sailing ships classed by LR over the last 10 years the vast majority employ seam welded, tapered, tubular sections.
Figure 19 shows the assembled sections of the topmast and topgallant masts being positioned on the lower mast section.
The rolling process limits the individual mast section lengths which can be accommodated and this results in a large number of circumferential butt welds. figure 19
In order for the masts to be structurally efficient some degree of control is required over the dimensional accuracy and build quality. Manufacturing tubular section from rolled plate and subsequent seam welding generally results in dimensional variation from the design values and the following criteria are recommended:
Nlì\iflhIifl1 variahofl Ifl -0.6t diameter Ma'imum ovalitv dx-d <=t
-:2t- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Where:- t= local mast wall thickness dx = outside diameter in the fore and aft direction dy = outside diameter in the transverse direction
Assembly of the individual rolled mast sections into complete mast assemblies requires control to ensure that successive sections have satisfactory alignment and that overall straightness of the mast is achieved. The recommended limits on these build parameters are:
lc1\lfltUtfl Eflts-dligflhìient t/ 4 between SUCCeSSLVC sections 5mm in a 4m length with a Niaximurn overall mast maximum of 20mm for the eccentricity from mast complete mast length. centreline
The completed mast butt and seam welds are required to be examined using an approved non-destructive test method. All butt and seam welds in masts are classed as 'critical welds' and the survey procedures of LR demand that the following non-destructive examination is carried out:
\Veki Category Butt Welds Fillet welds MPI US MPI Critical welds 100% 100% 100% Primary welds 100% 20% 100% Secondary welds 20% none 20%
At the time of the design appraisal of the masts and standing parts of the rigging, the locations and attachments of running rigging to the masts are generally not well defined. In many instances running rigging attachments are decided at ship once the masts, spars and standing rigging are installed.
-2- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Hence, as provision for these attachment points is carried out aloft, the primary method of attachment is by drilling and tapping the mast wall to accept bolted pad-eyes. Care has to be taken to ensure that large numbers of holes are not arranged in close proximity thereby locally reducing the structural integrity of the mast section. figure 20 Unlike the traditional sailing ships of the19thcentury, modern vessels utilise the internal mast cavity for the lead of electrical wiring to navigational aids and lights, hydraulic lines for the yard arm motors and in some cases for the running rigging. The exhaust piping and silencer from the auxiliary engine or generating setare often arranged inside mrzzen or jigger masts.
The masts also provide a convenient route for the air pipes of sewage tanks thus ensuring that venting is remote from the passengers and crew irrespective of wind direction. All components of the standing rigging are required to meet the following criteria:
Steel wire ropes used for the standing rigging are to be manufactured at an approved works and are to have manufacturers test certificates. Loose gear items such as shackles, bottle screws etc. are to be from approved manufactures and are to have certificates of proof testing.
Generally the shipbuilder will obtain the services of experienced rigging specialists for the preparation of the steel wire ropes with their terminal end connections. The proof testing of completed ropes depends entirely upon the type of end connections. Poured zinc or resin approved terminal end connections do not I , 4,L'Lho$&Iii4hor Li [L] rt,-rCt'1 fo4oALL,J Lt]4At%.,1for completion. Experience in the extensive use of this connection for lifting appliances has indicated that a high degree of security and reliability can be obtained with this type of terminal end connection. figure 22 THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
For swaged or talurit clamping terminal end connections the proof testing of the completed assembly has to date been carried out for each component. The proof test is not to be less than 40% of the nominal breaking strength of the rope. For traditional rope termination's such as clamping and seizing the termination can often only be completed at the ship with the rope in situ. Any prior testing of the termination only demonstrates the acceptability of the methodology rather than as an indication of the quality and overall reliability of the connection.
figure 21
Testing of completed ropes is a time consuming and expensive undertaking and where it is found that testing machines are unable to accommodate the long lengths of completed rope, methods of doubling ropes over large sheaves or clamping the rope have to be employed.
The vast majority of large sailing ships employ bottle screws to tension the standing rigging. The 'Stad Amsterdam' however, uses traditional deadeyes and lanyards for all shrouds. In place of hemp, the lanyards are of Dyneema and the dead eyes are of a high density polymer. The testing of the 'wall knot' is shown in figure 21 and the break testing of the completed assembly is shown in figure 22.
figure 23
-a- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
The pre-tension is required to guard against loss of tension in the leeward shrouds which in turn will subject the rigging to shock loads which will result in wear and fatigue damage to the components. With the vast number of lower shrouds, often 12 per mast, it is important that excessive pre-tension does not lead to load criticality for the mast column. The normal working load in the shrouds should not exceed 29% of the breaking strength of the rope (F.O.S3.5). Based upon the calculated shortening of the leeward shrouds from the finite element analyses undertaken, a pre.tension of no more than 15% of the breaking strength of the rope would appear to be sufficient to ensure positive tension at all times. Typically for a large barquentine with lower shrouds of 22mm diameter (6 x 36 GSWR) the pre-tension should not exceed about 4OkN. Tests with bottle screws and a suitable able bodied seaman would indicate that the maximum axial force which can in fact be generated is probably no more than 20 kN.
As part of the classification process, the scheme of pre-tension is required to be submitted and proposals are to be made regarding the methods of measuring and establishing that the intended pre-tension has been achieved.
All adjustable, demountable or removable rigging components or parts of components are to have mechanical securing methods to ensure that the effects of vibration, motion or load cycling will not result in loss of connection integrity.
A detailed rigging plan indicating the dimensions and construction of all standing rigging ropes together with a list of loose gear forms a very important reference document. As the rigging is surveyed, usually at annual intervals, correct identification and replacement of worn or corroded parts is essential for maintaining the safety and integrity of the rigging and more importantly the mast structures. THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
SURVEY AND SERVICE EXPERIENCE
An integral part of classification is the periodical surveys undertaken to ensure that the masts and standing rigging fulfil their function without compromise on safety. For LR classed sailing ships the schedule of periodical survey falls into two distinct categories;
The annual survey The quadrennial survey.
Due to the nature of the surveys, specialist rigging companies are employed to carry out the inspection, examination and reporting.
The content and level of survey is set out in the document
"Preliminary guidance information for the testing, marking, survey and certification requirements for masts and standing rigging."
The annual survey is concerned with the rigging arrangement, mast and rigging condition and the functionality and efficiency of all associated loose gear items. The annual survey is carried out with the masts and rigging in situ and in addition to a visual observation of the condition of the wires and fittings, the structural condition of the masts and bowsprit are reported upon.
The quadrennial survey is a more rigorous examination of the masts and rigging and normally requires elements of the standing rigging to be removed in order to gauge the amount of diminution or wear down of shackle and devis pins. Where masts exhibit corrosion or damage they are required to be un-shipped and thickness determination carried out using ultrasonic gauging techniques. At the time of preparation of this paper, the barquentines, Star Gipper fic!ure 24 and Star Flyer have been in service for almost 10 years and these ships have now undergone two quadrennial surveys.
-36- THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
The surveys and corresponding reports form a unique basis for verification of the adequacy of the initial criteria used in the design and verification process of the masts and rigging. The most commonly reported items in annual surveys are that the pre- tensioning of shrouds is not maintained or stays are too slack.
Sailing ships used for commercial enterprises such as passenger vessels may employ crews who are unable to maintain the rig or are not afforded the time or facilities to do so. In contrast to this, sail training ships would be expected to be maintained to a satisfactory standard as this function is an integral part of the vessel's purpose.
figure 25
Examples of items observed and noted during annual surveys of the standing rigging are shown in figures 24 and 25. Note, the mis-match between devis pin diameter and eyeplate hole in figure 25. A heavily corroded stay wire is shown in figure 27 and fouling of the running rigging may have been responsible for the loss of paint and galvanising from the wire in way of the terminal end in figure 28.
figure 26
37 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 49 Mastb of Sailing Vessels 1922Rules Imperial
Lengthof mast Diameter and thickness of mast Sizes of angle Cheeks
Fore and aft rig Square rig Partners Heel Hounds Head Bars In mast Thickness of plate Sizes of angle bar
feet feet inches inches inches inches inches inches inches inches inches inches inches
36 30 16 0.30 13 0.26 131/2 0.26 11 0.24 0.40 31/2 X21/2 X 0.34
37 31 17 0.30 131/2 0.26 14 0.26 111/2 0.24 0.40 3 1/2 X 3 X 0.36
38 32 i8 0.30 14 0.26 15 0.26 12 0.26 0.40 31/2 X 3 X 0.36
39 33 19 0.34 15 0.30 151/2 0.30 12 1/2 0.26 0.44 4 X 3 X 0.40
41 34 20 0.34 16 0.30 16 1/2 0.30 131/2 0.30 0.44 4 X 3 X 0.40
43 35 21 0.34 i6 1/2 0.30 171/2 0.30 14 0.30 0.44 4 X 3 X 0.40
45 36 22 0.36 17 0.30 i8 1/2 0.30 14 1/2 0.30 0.46 41/2 X3 X 0.40
47 37 23 0.36 18 0.30 19 0.30 15 1/2 0.30 0.46 41/2 X3 X 0.44
49 38 24 0.36 19 0.30 20 0.30 i6 0.30 0.46 4 1/2 X3 X 0.44
51 39 25 040 191/2 0.34 21 0.34 161/2 0.34 0.50 5 X 3 X 0.46
53 40 26 040 20 0.34 211/2 0.34 17 1/2 0.34 0.50 5 X 3 X 0.50
55 42 27 0.44 21 0.34 221/2 0.34 i8 0.34 0.50 5 X 31/2 X 0.50
57 44 28 0.44 22 0.36 23 0.36 i8 1/2 0.36 31/2 X 3 X 0.40 0.54 5 X 31/2 X 0.50
59 46 29 0.46 221/2 0.36 24 0.36 191/2 0.36 4 X 3 X 040 0.54 51/2 X4 X 0.54
62 48 30 0.46 23 0.40 25 0.40 20 0.36 4 X 3 X 0.44 0.56 6 x4 X 0.54
6 50 31 0.50 24 0.40 26 0.40 201/2 0.36 41/2 X 3 X 0.44 o.6o 6 x 4 X 0.56
68 52 32 0.50 25 0.40 261/2 0.40 21 0.36 5 X 3 X 0.46 0.60 6 X 4 X 0.56
71 54 33 0.50 z6 0.40 27 0.40 211/2 0.36 5 X 3 X 0.48 0.60 6 x4 X 0.56 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 49 Steel Bowsprits continued 1922 Rules Imperial
Length outside bed Bed Heel Cap Sizes of angle bars
DiameterThicknessDiameterThicknessDiameterThickness
feet inches inches inches inches inches inches inches
14 161/2 0.30 14 0.30 12 21/2 X 2 X 0.30
15 171/2 0.30 15 0.30 121/2 0.30 21/2 X 2 X 0.30
16 19 0.30 i6 0.30 13 0.30 3 X 2 X 0.30
17 20 0.34 17 0.34 14 0.30 3 X 2 X 0.30
18 21 1/2 0.36 18 0.34 15 0.30 3 X 2 1/2 X 0.30
19 28 0.36 19 0.34 16 0.30 3 X 3 X 0.32
20 24 1/2 0.40 20 0.36 16 1/2 0.32 3 1/2 X 3 X 0.34
21 25 1/2 0.40 21 0.36 171/2 0.32 3 1/2 X 3 X 0.36
22 261/2 0.40 22 0.36 18 1/2 0.32 4 X 3 X 0.40
23 28 0.44 23 0.40 19 0.34 4 X 3 1/2 X 0.40
24 29 0.44 24 0.40 20 0.34 4 X 3 1/2 X 0.40
25 30 0.46 25 0.40 21 0.36 41/2 X 3 1/2 X 0.42
26 31 1/2 0.46 26 0.40 21 1/2 0.36 4 1/2 X 3 1/2 X 0.44
27 33 0.46 27 0.40 22 0.36 4 1/2 X 3 1/2 X 0.46 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 49 Masts of Sailing Vessels 1922Rules Metric
Length of maat Diameter and t.hlckneaa of mast Sizes of angle Cheeks
Fore and aft rig Square rig Partners Heel Hounds Head Bara in mast Thickness of plate Sizes of angle bar
m m mm mm mm mm mm mm mm mm mm mm mm
x64 X 10.97 9.14 406 8 330 7 343 7 279 6 io 8g 9
10 89 X 76 X 11.28 945 432 8 343 7 356 7 292 6 9
10 89 X 76 z 11.58 9.75 457 8 356 7 381 7 305 7 9
11 102 X 76xio 11.89 io.o6 483 9 381 8 394 8 318 7
X 76 X 10 12.50 10.36 508 9 406 8 419 8 343 8 11 102
11 102 X76 X 10 13.11 10.67 53.3 9 419 8 445 8 356 8
X 76zlo 13.72 10.97 559 9 432 8 470 8 368 8 12 114
X 76 X 11 14.33 11.28 584 9 457 8 483 8 394 8 12 114
114 X 76 X 11 14.94 ii.8 6io 9 483 8 508 8 406 8 12
127 X 76 x 12 15.54 11.89 635 10 495 9 533 9 419 9 13
X 76 X 13 1615 12.19 660 10 508 9 546 9 445 9 13 127
16.76 12.80 686 11 533 9 572 9 457 9 13 127 X 8g X 1.3
17.37 13.41 711 11 559 9 584 9 470 9 89 X 76 X 10 14 127 X 89 X 13
17.98 14.02 737 12 572 9 6io 9 495 9 102 X76 X 10 14 140 X102 X 14
18.90 14.63 762 12 584 10 635 10 508 9 102 X76 X 11 14 152 X 102X 14
X 102X 14 19.81 15.24 787 13 610 10 660 10 521 9 114 X 76 X 11 15 152
20.73 813 13 635 10 6'3 10 533 9 127 X 76 X 12 15 152 X 102 X 14
X 102 X 14 21.64 16.46 838 13 660 10 686 10 546 9 127 X 76 X 12 15 152 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 49 Steel Bowsprits continued 1922 Rules Metric
Length outside bed Bed Heel Cap Sizes of angle bars
DiameterThicknessDiameterThicknessDiameterThickness
m mm mm mm mm mm mm mm
4.27 419 8 356 8 305 64 X 51 X 8
4.57 445 8 381 8 318 8 64 X 51 X 8
4.88 483 8 406 8 330 8 76 X 51 X 8
5.18 508 9 432 9 356 8 76 X 51 X 8
4.49 546 9 457 9 381 8 76 X 64 X 8
5.79 711 9 483 9 406 8 76 X 76x 8
6.io 622 10 508 9 419 8 89 X 76 X 9
6.40 648 10 533 9 445 8 89 X 76 X 9
6.71 673 10 559 9 470 8 102 X 76x 10
7.01 711 11 584 10 483 9 102 X 89 X 10
7.32 737 11 610 10 508 9 102 X 8g x 10
7.62 762 12 635 10 533 9 114 X 89 X 11
7.92 800 12 66o 10 546 9 114 X 89 X 11
8.23 838 12 686 10 559 9 114 X 89 X 12 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILINGVESSELS
Table 50 Yards and topmasta of sailing vessels 1922Rules Imperial
Yards Topmasta
Length Cleated Centre First quarter Second quarter Third quarter Ends at cleat Length Heel Lower part of head Head
ThIcknessDiameterThickness DiameterThickn saDiameterThicknessDiameterThicknessDiameterThicknessDiameterThickness DiameterThicknessDiameter
inches inches inches inches Inches feet juches jnchni inches inches inches inches inches Inches inches inches feet Inches
12 12 0.24 101/2 0.24 9 o.i8 32 8 0.18 77/8 o.i8 71/4 0.18 6 0.18 4 0.12
0.12 13 1/2 121/2 0.24 11 0.24 91/2 0.18 36 9 o.i8 83/4 o.i8 81/8 0.18 63/4 0.18 4 1/2
13 0.24 111/2 0.24 10 o.i8 40 10 0.20 93/4 0.20 9 0.18 71/2 0.18 5 0.12 15
51/2 0.12 16 1/2 0.26 121/2 0.24 10 1/2 0,20 44 11 0.21c 103/4 0,22 10 o.i8 8 1/4 o.i8 4
18 141/2 0.26 13 0.24 11 0.22 48 12 0.24 113/4 0.24 103/4 0.20 9 0.18 6 0.4
191/2 15 0.30 131/2 0.26 111/2 0.24 52 13 0.24 125/8 0.24 113/4 0.22 93/4 0.18 61/2 0.14
0.16 21 16 0.30 14 0.26 12 0.24 56 14 0,2f 13 /8 0.26 125/8 0.24 10 1/2 0.20 7
i6 1/2 0.30 141/2 0.26 121/2 0.26 6o 15 0.20 14 /8 0.26 131/2 0.26 111/4 0.22 71/2 oiS 221/2
24 17 0.34 15 0.30 13 0.30 64 16 0.30 155/8 0.30 143/8 0.30 12 0.24 8 o.i8
251/2 18 0.34 16 0.30 13 1/2 0.30 68 17 0.30 161/2 0.30 151/4 0.30 123/4 0.24 81/2 o.i8
27 i8 1/2 0.34 161/2 0.30 0.30 72 18 0.30 171/2 0.30 161/4 0.30 131/2 0.26 9 0.18 4
19 0.34 17 0.30 141/2 0.30 76 19 0.32 iB 1/2 0.30 171/8 0.30 141/4 0.26 9 1/2 0.20 281/2
iB 0.30 15 0.30 80 20 0.36 191/2 0.30 18 0.30 15 0.26 10 0.22 30 20 0.34
0.32 181/2 0.30 151/2 0.30 84 21 040 201/2 0.34 19 0.30 153/4 0.30 10 1/2 0.24 311/2 201/2
21 0.36 19 0.30 16 0.30 88 22 040 211/2 0.34 193/4 0.30 16 1/2 0.30 11 0.24 33
0.36 20 0.30 161/2 0.30 92 23 040 22 1/2 0.36 203/4 0.34 171/4 0.30 111/2 0.26 35 22
0.30 17 0.30 96 24 041 223/8 0.36 215/8 0.34 i8 0.30 12 0.26 37 23 0.36 21 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 50 Yards and topmasta of sailing vessels 1922Rules Metric
Yarda Topmasts
Length Cleated Centre First quarter Second quarter Third quarter Ends at cleat Length Heel Lower part of head Head
Diameter ThicknssDiameterThicknessDiameterThicknessDiameterThicknessDiameterThickness DiameterThicknessDiameterThicknessDiameterThickness
mm mm min mm mm ro mm mm mm mm mm mm mm mm mm mm m mm
3.66 305 6 267 6 229 5 9.75 203 5 200 5 184 5 152 5 102 3
4.11 318 6 279 6 241 5 1097 229 5 222 5 206 5 171 5 114 3
330 6 292 6 254 5 12.19 254 5 248 5 229 5 191 5 127 3 4.57
318 6 267 13.41 279 6 273 6 254 5 210 5 140 3 5.03 356 7 5
330 6 279 6 14.63 305 6 298 6 273 5 229 5 152 4 5.49 368 7
292 6 15.85 330 6 321 6 298 6 248 5 165 4 5.94 381 8 343 7
406 8 356 305 6 17.07 356 7 346 7 321 6 267 5 178 4 6.40 7
419 8 368 318 7 18.29 381 371 7 286 6 191 4 6.86 7
381 8 330 8 19.51 406 8 397 8 365 8 305 6 203 5 7.32 432 9
406 8 8 20.73 432 8 419 8 387 8 324 6 216 5 7.77 457 9 343
8.23 470 419 8 356 8 21.95 457 8 445 8 413 8 343 7 29 5 9 8.69 483 432 8 368 8 23.16 483 8 7o 8 435 8 362 7 241 5 9
6 9.14 508 8 381 8 24.38 508 9 495 8 457 8 381 7 254 9 457
6 9.60 521 470 8 394 8 25.60 533 10 521 9 483 8 400 8 267 9 io.o6 483 8 406 8 26.82 559 10 546 9 502 8 419 8 279 6 533 9
8 419 8 28.04 584 10 572 9 527 9 438 8 292 7 10.67 559 9 508
584 8 432 8 29.26 6io 10 568 9 549 9 457 8 305 7 ji.28 9 533 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 51 Standing iilng of sailing vessels 1922Rules Imperial
Secondlongltiidlnalnumeral 6200 7100 8000 9000 10000 11400 12800 14200
Lx( Il + D)
No Size No. Size No. Size No, Size No. Size No. Size No. Size No Size
inch inchea inchea inches inches inchee minea inches
Fore and main shrouds dr 4 of 2 1/2 4 of 2 3/4 5 of 5 of 31/4 5 of 31/2 5 of 3 3/4 6 of4 6 of 41/8 end can and cao and cap end cao and can
cham platea t i1/4 1 1/4 1 3/8 1 5/8 1 3/4 i3/4 1 7/8 1 7/8
dead eyes dia xt 7 X 4 i/a 7112 X 41/2 8 o 5 81/2X 5 9 051/2 91/2X 5 1/2 io x 6 101/2 X
lanyardo (hemp or 3 1/2 3 3/4 4 41/4 4i/a 43/4 5 51/4
rigging screwa, diameter at bottom of thread cha is/S 1 i/S 11/4 1 3/8 11/2 11/2 i 5/8 1 5/8
rigging acrews diameter of pino dia 1 1 1 i/8 11/4 1 5/8 13/8 s /8 1 3/8
topmestbksti,ya dr 2 of 21/2 0 of 2 3/4 2 of 3 1 of 3 5/4 a of 31/2 2 of 3 3/4 of4 of4i/8
top-gallant baci otays dr 1 /4 2 21/8 2 i/4 2 3/8 a a s/a 2 2 5/8 2 2 3/4
lower stays or 2 of 21/2 2 of 2 3/4 2 of 3 2 of 3 1/4 1 aS 31/a 2 3 3/4 2 4 2 4 1/8
topmaßt stays nr 21/2 2 3/4 3 3 1/4 2 31/2 2 3 3/4 2 4 2 -4 1/8
top-gallant sta3ii dr 1 3/4 2 a 1/8 2 1/4 2 3/8 2 1/a 25/8 2 3/4
Mimen shrouds dr of 21/4 3 of 2 3/8 4 of a s/a 4 of a 5/8 5 of 23/4 5 of 2 7/8 5 Xi 3 5 of cori rar,
topmast bachstays dr a1/4 2 3/8 a o/a a -a 5/8 2 2 3/4 2 -2 7/8 3 3 3 /4
top-gallant beckstays or 1 1/4 1 3/8 i1/2 i5/8 1 3/4 1 7/8 a a a 21/8
lower stays or 21/4 2 J8 2 s/s a 5/8 23/4 2 7/8 2 3 2 3 1/4
topmast stays dr 21/4 23/8 2 1/2 2 5/8 2 3/4 2 7/8 3 2 3 5/4
top-gallant stays dr s1/4 i3/8 1 1/2 1 5/8 i3/4 1 7/8 2 2 5/8
Bobstay bar dia 2 2 2 2 21/4 a o/a 31/4
pin die 11/2 1 1/2 1i/s 1 1/2 1 5/8 07/8 21/8 2 1/4
chain dia i3/16 1 1/4 1 1/4 i1/4 1 5/16 i3/8 1 1/2 1 5/8
Boweprot abrouds (chain) cia g/iS 9/16 5/8 u/ii 3/4 a of 3/4 2 of 1.3/16 2 of 7/8 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 51Stan.ling rigging of sailing vessels continued 1922Rules Imperial
Secon.I longitudinal numerai 15600 17000 18400 20000 21900 24200 27200
Lx( B + D)
Size No. Size No, Size No, Size No. Size No. Size No. Size Ño.
inches Inches inches inches lochai Inchei inches
Fore and main shrouda tir of 41/4 6 of 4 1/2 6 of 43/4 6 Of 7/8 6 of 5 6 Of 5 1/4 6 of 5 i/a end cao and ceo and cao end Cao and can aniS Cal) champisses t a ai/S 21/4 a3/8 2 i/a a5/8 23/
deed e es dia xi ii X 6 n i/o X 6 1/2 12 X 7 - - - -
lanyard s (hemp) nr i/a 5 3/4 6 - - - -
2i/B 2 1/4 riggln screws diametor at botthm of thread dia 13/4 13/4 1 7/8 1 7/8 1
1 7/8 a ri.ggin screws diameter of pins dia 1 1/2 1 1/2 1 5/8 1 /8 1 3/4
topmeotbackstays tir of 4 1/4 3 of 4i/a 3 of 43/4 3 Of 4 7/8 3 of 5 3 Of 5 i/. 3 of 5 i/a
2 2 top-ga iantbackstays tir a 3 2 31/4 2 3 1/2 a 33/4 2 3 7/8 4 i/S 4 1/4
a 1 lower .ays tir 2 4 1/4 41/2 2 4 3/4 2 47/8 2 5 5 1/4 5 1/2
2 a 5 1/4 2 5 1/2 topmat stays tir a 4 1/4 2 41/2 2 4 3/4 2 47/8 5
1/8 i/. top-gallant stays tir 3 31/4 3 1/2 33/4 3 7/8 4 4
1/8 1/4 of 4/8 of 41/2 Mizzen shrouds cii 5 of 3 1/2 5 of 35/4 5 of 4 5 Of 4 5 of 4 5 5 endran endren endran sudran andren andres endest /8 1/2 topmast backainys dr 3 3i/a 3 33/4 3 4 3 -4 1/8 3 4 1/4 3 4 3 4
a top-gallant backstaya tir 2 21/4 2 -ai/a z a /4 2 2 7/8 2 3 a 3i/B
2 1/4 2 /8 2 1/2 lower stsys nr a 31/2 2 3 3/4 2 -4 2 41/8 - 4 4 4
1/4 2 2 4 1/2 topmast stays tir a -3 i/a a -33/4 2 - 4 2 4 o/8 a 4 4 3/8
1/4 top-gallant atayo tir 2 1/4 2i/a 2 3/4 2 7/8 3 i/B 3
41/8 4 1/8 Bobstay bar dii 3 1/a 33/4 3 3/4 5 7/8 4
1/8 pin dia 2i/a a5/8 2 3/4 2 7/8 3 5 3 1/8
a a i/iS 21/16 chain dia 1 3/4 in/io i7/8 1 1.5/16
2 a ofi 1/16 a ofi1/16 1 of1 1/8 2 ofii/B Bowsprit shrouds (chian) dia a of 7/8 2 of1 ofi HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 51Standing rigging of sailing vessels 1922Rules Metric
Second IongItidina numer& 576 66o 744 837 930 io6o 1190 1320 Lx(l3+D)
No. Size No. Size No. Size No. Size No Size No. Size No. Size No. Size
mm mm mm min mm mm mitI mm
of 30 6 of 32 6 of Fore and main ahroudg dia 4 of oo 4 of 22 5 of 24 5 cl 26 5 of 28 5 31 ad cat) artd andjj an. COIl chain plateo t 32 32 15 41 44 44 48 48
X 229 X 140 241 5 140 254 X 152 267 a 152 dead eyes dia X t 178 11 U4 191 X 124 203 X 127 216 127
lanyarda (hemp) dia 28 30 32 14 36 38 40 42
rig1ingacrewedthineteratbottomofthread dia 29 29 32 35 38 38 41 41
rigginI screws, .iiaineter of pins dia 25 25 29 32 41 35 35 35
topmast backstt'a dia 2 of 20 2 of 22 2 of 24 2 of 26 2 of 28 2 of 30 3 of 32 3 of 33
top-gailantbacktays dia 14 16 17 1.8 19 2 20 2 - 21 2 22
lower stays die 2 of 20 2 of 22 3 of 24 2 of 26 2 of 28 2 30 2 32 2 33
topmast stays dIa 20 22 24 26 2 28 2 30 2 - 32 2 33
top-gllantstas dio 14 1 17 i8 19 20 21 22
24 of 26 Miueñ shrouds dia 3 of i8 of 19 4 of 20 4 it 21 5 of 22 5 of 23 5 of 5 end rar
topmast bacistays dia 18 19 20 2 21 2 22 2 23 3 - 24 3 26
top-gallant backotays dia 10 11 12 13 14 15 2 i6 2 17
lower aleya dia i8 19 20 21 22 23 2 24 2 26
topmast otaya dia 18 19 20 21 22 23 24 2 26
top-gallant stays dia io u 12 13 14 15 16 17
Bobatey bar dia 51 51 51 51 57 64 76 83
pin dis 38 38 38 38 41 48 54 57
chain dia 30 32 32 32 33 35 38 41
Bowsprit ahrouds (chain) dia 14 14 16 17 19 2 of 19 2 of 21 2 of 22 HE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
Table 51 Stan cling rigging of sailing vessels continued 1922Rules Metric
2528 Secor IongtudInaI numeral 8450 1580 1710 1859 2036 2249 Lx(B+D)
No. Size No. Size No. Size No. Size No. Size No. Size No. Size
mm mm mm mm mm mm mm
8 of 42 6 of Fore and main ahroud dia 6 of 34 6 of 96 6 of 38 6 of 3 6 of 40 44 end aod and caz and ran andi nd 70 chain platea t 51 54 57 60 64 67
- - dead .,'ee dia xi 279 X 152 292 X 165 305 X 178 - -
lanyaili (hemp) dia 44 46 49 - - - -
rigginscrewa,diameterotbottomofthread dia 44 44 48 48 52 54 57
48 51 rigsiriscrews, diameter of puis dia 38 38 41 41 44
42 of topmitbackatays dia 3 of 34 3 Of 36 3 of 38 3 of 39 3 of 40 3 of 3 44
top-gi!lantbackstaya dia 2 24 2 26 2 28 2 30 2 31 2 33 2 34
lowec itaya dia 2 34 2 6 2 38 2 39 2 40 2 42 2 44
2 2 40 2 42 3 topm.lit atays dia 3 34 2 36 2 38 - 39 44
top-gillant stays dia 24 36 28 30 31 33 34
Of of of 36 Minen shrouda dia s of a8 5 Of 30 5 of 33 5 rif 33 s 34 5 35 5 . . -Ial. .1.110.I .01* . . . .
- 35 3 36 topmastbkiisya dia 3 - 28 3 30 3 32 3 33 3 34 3
top-gallantbakataya dia 2 i8 2 - 20 2 22 2 23 2 - 24 2 25 2 26
2 2 96 lower stays dia 2 28 2 30 2 " 32 2 33 3 34 35
2 36 topmast staya dia 2 28 2 30 2 - 32 2 - 33 2 34 2 - 35
top-gallant sta dia 18 20 22 23 24 25 26
105 105 Bobatay bar dia 89 95 95 98 ins
pin dia 54 67 70 73 76 79 79
52 53 chain dIa 44 46 48 49 51 29 Bowsprit shrouda (chi.in) dia z of 22 2 of 25 2 of 25 2 of 27 2 of 27 2 of 39 2 of THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS
References
Thi s per s p epar ed w t h t he hd p and assi st ance of a nuiter ofi ncvi dual s and Organi sat i ons. FE anysi s exarç4 es and dat a on nany sai li ng shi vere olai ned f r orn depar nent s of LI's' d Rçj st er' s I-adquar t er si n Lancbn. flcÍ o ai c nat er i al and report s on sur veis ver e p ovi dad bj IR s su veyor s andt o a I ar ext ant,f r orn THE VERIFICATION OF MASTS AND RIGGING OF LARGE SAILING VESSELS the specialist rigng organisations viio in nariy cases, desigi, erect, survey aid nai ntain large sailing ships vcrldÑde.
Part i cul ar nrit i on i s nade of t hose desi glers, U c'cf s Fèçj st er has vorked wt hover t he I ast decade vJio have vil Ii nç y sha-ed t hei r kncwk edge and pert i se on t hi s sped alised topC.
Ust of contributors:
(rard Djkstra &Ftriers val Pr chit ed s and F'k i ne Bineer s
Ctean ii ng Evel openent I-blI and Bi \i I3er I e I raat 10 Pnst er dam 1071 4W
vccn. naval consul t i ng Gt,t-j \M gast Esi gaers, surveyors, consLi tangs and riggers of tal Iships Bienstrasse 58 D22765 l-biturg
thoren Cèsi & nsuI t i ng Moindsena r. SA R. 80-250 )*9<44 PO Bx31, R1W
60- PRACTICAL EXPERIENCE ON REDUCING MOTIONS AND IMPROVING COMFORT ON BOARD LARGE MOTORYACHTS
Ir. H.M. van Wieringen De Voogt Ship Design B.V.
Ing. F.A. Gumbs De Voogt Ship Design B.V.
Ir. R.P. Dallinga MARiN
INTRODUCTION
In the course of 1996 a very detailed investigation was carried out in order to improve the passenger comfort on board large motoryachts. The l4' symposium on "Yacht Design and Yacht Construction" in 1996 contains the paper on this subject. Also at Project '98 a paper has been presented on "Improving Motion Comfort on Motor Yachts". For all these investigations and studies it is imperative to have a set of 'comfort criteria' to judge against. As literature provided only limited data on criteria applicable for motoryachts, two methods were explored to improve on the knowledge on comfort. First a set of tests was performed in a motion simulator at TNO Zeist, involving many test-persons to judge on the comfort or dis-comfort of various type of motions. Second a long-term measurement program was organised and performed by MARIN, comprising actual motion measurements on board two motoryachts and the simultaneously gathered comfort ratings of the crew and passengers on board.
SH[P AND SH[P MOTION CHARACTERISTICS
For each yacht a small-scale body plan is attached in figure 1 and figure 2
The figures show, from a hydrodynamic perspective, two very similar ships. The 40.92 m Carmac VII has a displacement of 436 Tf. The 0.85 m GM yields in combinations with the initially adopted transverse radius of inertia of 2.8 m (32.4% B) a natural period of roll of around 6.7 s. The 41.0 mEnterprise V has a displacement of 465 Tf. The slightly higher 1.03 m GM yields in combination with the initially adopted transverse radius of inertia of 2.8 m (32.4 % B) a slightly lower natural period of roll around 6.1 s.
For both vessels the rolling characteristics are very similar. The response is characterised by a sharp resonance peak around the natural frequency of roll. Noteworthy is the fact that the highest response is obtained in bow- and stern- quartering waves. A consequence is that the vessel requires a very careful alignment with the wave direction to reduce the roll motions: a heading 22.5 degrees off head seas already yields half the maximum response. The pitch angles of both vessels are much lower than the roll angles. Both vessels show the highest response in a broad range of headings around head seas with values exceeding 4 deg!m. The yaw response is the lowest of the three angular motions; it ranges around 2 deglm in bow- and stem-quartering waves.
Considering the vertical accelerations it is observed that at the bow and the stem the pitch response may be recognised to some extent. In beam seas heave dominates the vertical motions and accelerations. Considering a reference point at 30 m from the stem as typical for the motions in the forward half of the vessel, the results show that the vertical accelerations are not very susceptible to the wave direction. The results in head seas are only marginally lower than in quartering and beam waves.
Contrary to the vertical accelerations the effect of the longitudinal position on the transverse accelerations is rather small. This may be understood by considering the dominant contribution of the roll in the transverse acceleration levels, a value, which is the same over the length of the vessel (at the same height).
COMFORT CRITERIA
Because MARIN is trying to predict a complete picture of the performance of ships and structures under operational conditions, literature describing ship-motion related human performance reductions is followed with interest. Considering in-house information in the light of the above comfort concept it must be concluded that the existing literature focuses primarily on discomfort and task related biomechanical problems.
An aspect which is not covered by the above is the functionality of the ship as an 'entertainment platform'. These aspects are highly dependent on the anticipated human activity. They range from the feasibility of transfer of guests to smaller boats, feasibility of activities in and around the water and the feasibility of delicate on-board activities like formal dinners. The most frequently discussed discomfort issues are the incidence of seasicimess and human mobility (standing, walking). Intellectual performance in a complex technical environment under the influence of motion induced fatigue seems an emerging area. Motion Sickness Incidence (MSI) due to single narrow banded vertical motions is a relatively well-developed area. Important uncertainties within the present contect are the effect of combined motions and habituation in a varying environment. Also it is not clear up to which extent the MSI estimate is a relevant measure for passenger discomfort. Noteworthy is the fact that the range of accelerations in which MSI is low resides around 0.2
TNO simulator experiments
The motion simulator offered the opportunity to perform experiments in which the relative magnitude of various motion components could be varied systematically. Experiments during which typical "yacht activities" take place give insight in the 52- relative importance of these components and thus in effective ways to improve yacht design. The simulator experiments focussed on intellectual tasks (memorising numbers, reading and memorising letters while being subjected to considerable head movements), maintaining balance (walking a straight line and building a cube tower) and a game (throwing darts). These experiments were performed with 8 subject couples under 4 levels of roll (corresponding with 0, 0.05, 0.10 and 0.15 mIs2 transverse accelerations) with and without a 0.05 mIs2 rms heave motion. In all conditions a pitch motion with and rms of 0.5 degrees was added. In addition coffee, lunch and tea breaks were spent on board the simulator with slowly increasing motion amplitude.
The quantitative scores as well s the task related and general personal impressions of the participants were subjected to a statistical analysis to identify the significance of the various effects.
RESULTS
A very clear result from the simulator experiments is the fact that the transverse accelerations have a significant effect on the perceived effort and on the related scores. A transverse acceleration level of 0.1mIs2 rms proved to be half way on the 5 point scale between 'not at all' and 'extreme' influence of the motions on the effort to complete the task. Roughly the same result is obtained from the general questionaire when considering a point halfway acceptable and unacceptable on a 5 point scale. The number of MLI's during walking yielded a very straightforward relation with the rms transverse accelerations; it increased from O per 6 minutes for the case without transverse accelerations to i per minute at 0.15 mIs2 rrns. Another very clear result is the fact that the presence of the 0.05 mIs2 rms vertical accelerations is of very little influence on the foregoing results.
Timing the events during the lunch and coffee breaks showed that around 50% of the subjects started commenting on the motions at an rms level of around 0.1 mIs2 at 0.14 mIs2 50% were of the opinion that the motions were unacceptable.
RESULTS OF THE ON-BOARD MEASUREMENTS
Data
The records cover in total two three-month periods on board the two yachts.
Figure 3 shows a sample time history of the data statistics. From each "reported" comfort rating (Figure 4) statistics are available of the roll response and the vertical and transverse accelerations at key locations. The volume of the data in the various categories is indicated in Figure 5. Figure 6 indicates the correlation between the recorded roll motion and the recorded transverse acceleration. A very good coherence is obtained; indicating that roll is a prime factor in the transverse accelerations. The lower values in the cloud correspond closely with the inevitable gravity component along the deck. The dominance of the roll suggests that sway and yaw contributions are relatively small; this implies that by reducing the roll response the acceleration levels can be reduced as well. Figures 7 and 8 indicate the trend of the vertical and transverse accelerations with the comfort ratings. It shows that in the case of the vertical accelerations there is no clear relations between the comfort rating and the roll motion a very clear and similar trend is obtained. These values are summarised in the following table.
Failed Very bad Bad Good Very good Rms Tr.Acc. mis2Mean 0.24 0.20 0.13 0.09 0.09 Median 0.26 0.19 0.09 0.07 0.06 Rms Roll deg Mean 1.10 0.95 0.56 0.38 0.28 Median 1.48 0.91 0.44 0.26 0.14
Taking the median value as the most robust estimate the transition between bad and good is around 0.08 mIs2. A value, which is quite close to the 0.1 mIs2 suggested by the motion simulator trials. The value for very bad (0.19 mIs2 rms) is close to the 0.14 mis2 suggested by the coffee-break and lunch trials on board the motion simulator. One obvious characteristic of the present data is that the rms values can not take negative values. The fact that the Rayleigh distibution reflects this characteristic and the reasonable fit with the measured data motivated its use to predict 10% and 90% exceedance values.
Ragin level Very Bad & Failed Bad Good Very Good Rms Tr. Acc. 10% 0.065 0.051 0.035 0.038 mIs2 50% 0.167 0.131 0.090 0.096 90% 0.304 0.240 0.164 0.175 Rms Roll deg 10% 0.34 0.23 0.17 0.16 50% 0.88 0.59 0.44 0.42 90% 1.60 1.07 0.80 0.76
Comparing the median values from the fitted Rayleigh distribution with the median value from the data a reasonable agreement is obtained.
DISCOMFORT CRITERIA AND DESIGN IMPLICATIONS
Discomfort criteria
The reported comfort ratings correspond with a volume of accelerations statistics. These statistics show a considerable scatter, which indicates that the issue of discomfort is only partly understood. Despite the scatter, the correlation between the median values and the comfort ratings suggest strongly that the transverse accelerations are a prime indicator for passenger discomfort. These median values, which correspond with a 50% score, suggest that an rms transverse acceleration around 0.1mIs2 is the level in which in 50% of the reported cased the motions are characterised as "bad". lt is encouraging to note that the on board measurements and the Motion Simulator Trials in schematised conditions yield very similar levels. The results of the on-board measurements as well as the Motion SimulatorTrials agree also on the result that the verticalaccelerations are only of minor influence (within the range of the present data set).
Design implications
From literature the following five 'human factors engineeringprinciples' to optimise human performance in ship operations and design are suggested: locate critical stations near the ship's effective centre of rotation minimize head movements align an operator with the principal axis of the ship's hull avoid combining provocative sources provide an external frame of reference
In the light of the present results the remark on combined sourcesin terms of combined motions bay be less relevant. The same is true for thelongitudinal location on board the ship; sincevertical accelerations are less important the locations w.r.t. to the pitch centre is also less important. The present results do suggest that the emphasis on reducingroll is important to improve yacht design. As outlined in earlier work on behalf of Feadship measures to increase the roll damping are quite effective. On-board recognitionof the actual wave conditon (wind sea, swell) and direction by means of ship radar andsubsequent evaluation of the optimum heading on basis of the known motioncharacteristics seem feasible with contemporary radar and computer technology. By meansof a stern thruster or increasing the wind-vaning tendency of the vessel it mightbe easier to maintain the optimum heading.
CONCLUSIONS
In the course of 1997 and 1998 the motions of two motor yachts werecontinuously recorded. The results of measurements were related to questionaire basedcomfort ratings. The results were given an interpretation in the lightof dedicated Motion Simulator Trials by TNO and existing literature on human factors.
Based on the results it seems justified to conclude that:
The returned comfort rating shows considerable scatter which suggeststhat the rating is both rather subjective and situational. Both the on-board measurements as weH as the simulator trials suggestthat the transverse accelerations are a relevant measure fordiscomfort. Both data sets also suggest that within the present data sets the vertical accelerations are less important than the transverse accelerations. The data suggest that an rms tamsverse acceleration of 0.1mIs2 will be rated as 'bad' by 50% of the passengers. This value is considerable lower thansuggested by the existing literature.
Based on these fmdings it seems right to conclude that the ongoingefforts to reduce the roll motions of the present class of ships is very relevant todesign more comfortable motor yachts. Suggestions for improvements in yacht operations were made. - ENTERPRISE V
TENDER OPERA11ONS
\ SENSORBOX
NTEFRi V
_____J o -- s s s
RELAXATION EATiNG AND DRINKING
iiL41üi!j __ i ; lE CARMAC VII TENDER OPERATIONS SENSOROX
- ---
MAIN LOUNGE EATING AND DRINKING RELAXATION Carmac VII: July 1997
J -
2 II
I I JI'L
/ I IL I Il I L i' 'MH
L I II I ILil LI L
L __ - -.-_ --- - -4L ______,-1 I L I L L) J LI Li I o L j I L L II I I L II LI L L f\ tL1 L I LI IL \I I) III I LVi IL 1L L L I ill I Li IL I L
L I, i ';,V, L
-2 I
-3
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Day
Accy Std Accy Min Accy Max XAccy Marker Figure 4
HADSHIP
Captain's Questionnaire Comfort at Anchcr
M.Y. Respondent frl E-P'T4- Date Position or location - 7t/7 Number of guests
Local time: ,2c'c Greenwich Mean Time:'2-.oO Number of sick guests: NIL-. The relevant event was: tender operations & water sports, D eating & drinking; galley, D sleeping & relaxation, D other, namely: My judgement on that event is: ,,- D failed D verybad bad D good D very good Comments ¡flv î Fe . IrJC...7ijti¡.SS. IPÇ. H Localtime: /330. Greenwich MeanTime:z3ac Number of sick guests: j The relevant event was: D tender operations & water sports, Cil'eating & drinking; galley, D sleeping & relaxation, D other, namely:
My judgement on that event is: D failed D verybad [] bad D good D verygood Comments
'C 3EY\ CT
Local time: ,'jj cO Greenwich Mean Time: Number of sick guests: 2' The relevant event was: Li tender operations & water sports, D eating & drinking; galley, D sleeping & relaxation, D other, namely:
My judgement on that event is: ( failed D very bad D bad D good D very good Comments
. cS,N
My overall judgement of the day / D very bad C bad D good D very good Comments - - No Data Entries
120-
100
80- Number 60- Enterprise 40- Carmac 20- LII Total
0- failedvery bad bad good very good Comfort Rating Relation between Rolland TransverseAccelerations
0.6000
,_0.5000 CN U) 0.4000
0.3000
0.2000 E L i 0.1000
0.0000 0.0000 0.5000 1.0000 1.5000 2.0000 2.5000 3.0000 rms Roll [deg] Comfort Rating & Vertical Accelerations
0.3000
0.2000- rms Tr. Acce [rn/s 2] 0.1000- av mean Uav median
0.0000- failed very bad good very bad good Comfort Rating Comfort Rating & TransverseAccelerations
0.30
0.25
0.20 rms Tr. Accel. [mIs2J 0.10 at mean Uat median 0.05
0.o0 failedvery bad bad good very good Comfort Rating
"SOME CRETIG4 L J\TO TE S U DFSIGNJG WITH COMPOSITES"
Jons Degrieck Ghent University, Belgium Department of Mechanical Construction and Production
Abstract Composite materials are being used for some time now as construction materialsin nearly all domains of industry. Jndeed fibre reinthrced plastics allow for potentially high benefitsniosily on basis of their high specific stiffness and strength. Glass, carbon and aramidfibres are the most important candidates for plastic reinforcement. Very often however, one sees that well suited applications do not show theexpected results, or even fail early or in nn unexpected nnd catastrophic way. Diie Io theinherent nature uf libre reinthreed composites -of non-homogeneous, anisotropic and generallylayered materials, with practically speaking no plasticity- designing with composites is much differentfrom designing with the more classical construction materials and should be done with the necessary care. This contribution deals with a few critical flotes on designing withcomposites in yacht construction. A recent case of problems encountered with the snapping offof a 27-meter carbon fibre reinforced mast will be used as more elaborated case study.
Introduction
Composite materials are being used for some time now as constructionmaterials in nearly all domains of industry. This is also the case in yacht and boat construction.Indeed fibre reinforced plastics allow for potentially high benefits, mostly on basis of their highspecific stiffness and strength. Glass, carbon and aramid fibres are the most important candidatesfor plastic reinforcement. They can be combined with wood, foam or honeycomb cores to obtain lightand stiff panels for deck, hull and partition construction.
Very often however, one sees that well suited applications do notshow the expected results, or fail early or in an unexpected and catastrophic way, even ifthey do comply with all standards and regulations.Due totheinherent nature of fibrereinforced composites -of non-homogeneous, anisotropic and generally layered materials, with practicallyspeaking no plasticity- designing with composites is much different from designing with the moreclassical construction materials. As a result the actual composite may differ from the compositein the mind of the designer. Besides complying with prescribed standards and regulations, designingwith composites asks for "good composites practice" and "good composites workmanship".
More than with metals, the fabrication processshould be taken into account already at the design phase, as the structura! properues are detenmnncd loi onlyby the material eoittiiucnLs -the fibres and the matrix-, but also, and to a large extent, by thefabrication process. Close collaboration between designer and manufacturer is necessary at all stages of thedesign and fabrication process.
Of course this short contribution can not be meant as a full course ondesigning with composites, but simply wants io deal with a few critical notes on designingwith composites in yachi construction. Aller mentioning two very common problems in yachtconstruction, the recent case of problemns encountered with the snapping off of a 27-meter carbonfibre reinforced mast will be used as more elaborated case study.
- Sandwich structures
Sandwich structures are used extensively in composite yacht design. As wellknown, they allow obtaining light sm'ctural piinels with very Júgh bending stiffness and high bending strength.They typically consist of two outer skins from fibre reinforced plastic. and a core of balsa wood.organic foam or eventually Nomex or aluminium honeycomb. Their use ranges fromdeck snrliices, lo complete hull sections. internal partitions and stiffeners and interior panels. hi sandwich panels, bending moments are nearly fully worn by the skins,while transverse shear forces are taken mainly by the core material. Especially this last aspectasks for the attention of designer and manufacturer: correct transmission of transverse shear forces meansthat there must not be any gaps in the core material. Good bonding between the typicalsmall balsa blocks is really necessary, as well as between thecnerally larger panels in ihe 'ase of a foam or a honeycomb core (Figure 1). Failing to do so will result in shear forces having to be transmittedlocally by the skins bridging the gaps, and will hence result in local high three-dimensional stress componentsbetween the skins and the core. The associated out-of-plane components are oftenthe reason for early and unexpected opsetandgrowthofskin delamination
ft
Figure 1: In order to assure a correct transmissionoftransverse shear jbrces, carefully bondingof'adjacent care sections is necessary.
A second point of attention is the protection of the core material. Afterhaving trimmed a sandwich panel, or after having placed a screw Sor example (Figure 2), the corematerial should be carefully protected from water.Especially with wooden and honeycomb cores, wet core problems are frequently found on composite yachts. In case of (balsa) wood swelling and rottingis obvious with aramid (Nomex) cores swelling due to water pick-up is the main problem,whereas for a carbon fibre skinlaluminium iioueycomb core, corrosion is io be exp ectL In iii! cascsthe laniinate looses its nature of a sandwich structure, leading to a general loss of structuralintegrity. Once more or less large amounts of waler Jmve got into the sandwich core, freezingfurther accelerates the problent Willi closed cell foams, the problem is generally smaller.
t
I.
Figure 2: AjIer having trimmed a sandwich panel or having put a screw, the core material should be carefully protected from water.
Finally, one should be very careful with applying loads perpendicular to asandwich panel. as is most often the case when screwing attachments. Even in the case of apparentlysmall loads problems may have to be expected, because of the low bending strength of the relativelythin skins. Local metallic
- inserts or the use of stronger core materials (such as hardwood) is advisable in these areas. Where large loads have to be applied local skin thickening or even a complete different design should be adopted.
Bolted joints
As a second topic, the oflen-problematic use of bolted joints is mentioned. Bolted joints are frequently used for fixating metallic mountiig brackets. In bolting metallic parts. loads are generally transferred by friction forces between them (Figure 3,a); the bolts only have to guarantee the pretension that is necessary to assure .adequate friction. When connecting composite/composite or composite/.memllic components however, keeping the necessary pretension may be difficult, due to stress relaxation in the composite load transfer by solely friction is no longer possible, and the joint will become loose. In such cases the method of the Figure b is advisable, where most of the load is transferred directly through the bolt. In these cases holes have to be drilled with The greatest care and clearances should he kept as low as possible.
a) b)
Figure 3: Typical bolted joints for metallic parts (a), and for composite/composite or composite/metallic components.
A case study: snapping off of a 27-meter carbon fibre reinforced mast
Background As a third example the early snapping off of the mast of a high performance sailing cmiser is used here as a more elaborated case study io demonstrate the necessity of a careful design. One does not need to emphasize the role of the mast for a sailing yacht. Nowadays the masts of performance sailing cruisers are made relatively frequently from carbon fibre reinforced composites. Using bigh modulus carbon fibre, very stiff masts can be obtained at lower weights. This is very advantageous in reducing the necessary keel weight for example.
During summer 1999, the mast of the 62-feet sailing yacht "Pollux" did snap off during its maiden thp. after about 300 miles. The captain was a well-experienced skipper and weather conditions were tiormal. On request of its owner, the Department of Meciìaiucal Construction and Producuoii of Gheiit University has made a thorough study towards the possible cause(s) of this early mast failure. Laminate inspections were performed and a large-scale experimental set-up was conceived in order to reproduce as close as possible The seal mast situatiorL After niast replacement stain gauge measurements were performed on the lower mast rigging in order to verify rigging and mast loads resulting from mast pretension. Because the additional forces due to seal sailing conditions are not always well known, it was finally decided to perform measurements while sailing at sea. The Figure 4 gives a general view of the original 27-meter mast. as well as of the sails, the spreaders. and the diagonal and vertical rigging rods.
Figure 4: Genera! view of mast, spreaders, stays and saiLs of the 62' "Polluf'su.iling yachi.
The Figure 5 shows the damage observed at the so-called "Dl"-stage, where the lower diagonal rods are connected to the mast by means of a nieta! micro-tang, as shown in Figure 6. During a meeting between all parties involved, it became clear that mast failure probably had its origin at this Dl-stage. lt was then decided to perform destructive lests to estimate the actual bearing strength of the composite/micro-tang combination at the "D1"-stage. A non-damaged section of the mast, containing the original "D2"-stage was selected for this purpose. Allowing to the original lamination drawings, the "D2"-stage was nearly identical to the "Dl"-stage. Two additional tests were performed on another test specimen cut from a higher non-damaged section Df the mast. Holes were drilled near both ends of this section for insertion of the micro-tang bushes. Although the laminate in this section did not include the additional diagonal patching (of the original "D"-stages), test results gave a good indication of the bearing capability, as the additional patching at the "D" -stages was very low.
The test set-up An "ad hoc" test set-up was conceived so as to be an imitation, as faithfùl as possible, of the real situation of the mast on the ship. including mast prc-srtessing. The ncrmal mast pro-bending which results from pre-stressing however could not be simulated because of the short mast specimen available prohibiting the application of the necessary bending moments. It was estimated however that this only has a minor effect on the resulting bearing power of the composite/micro-tang combination. Figure 5: View of the danwge of the "Pollux" mast at the "Dl "-stage.
aM w,-.., TO Mf STW
Figure 6: Micro-tang with rods and bearing bushes; design of cross-sections at the "D "-stages.
The Figure 7 and Figure 8 give an upper schematic view, respectively general view of thetest set-up. The test section was put in a horizontal position with the forward face towards the floor (fullbatten rail upwards). Hydraulic cylinders provided the axial pre-stressing (further called Faxiai).as well as the forces of the diagonal rods on both sides of the micro-tang (Fnrther c.al!ed FF and The were placed under an angle, relative to the mast axis, corresponding to the real situation.
Applied load histories and e.xperimental results The applied load histories are depicted in Figure 9. Ina first step mast pre-stressing was simulated: it consisted of gradually applying forces, up to 13 0-140 kN nominallyon the mast itself (Fi). and up to 30 kN nominally on the micro-tang rods (F andFt0d).In a second step. wind force was (statically) simulated: thereto, the force on the rod at port side was gradually increased while the other
69 two forces were kept constant (within reasonable limits).It was decided to go up to the bearing strength of the composite/micro-tang combination.
00 BO VE NA AN ZCH T
00 Hydrautisehe zuiger cretan iO O Stuurbaord .-. Mosisteun voet O MAST
Mosisteur t 20 o 00 Hydraulische zuiger 00 00 Bakboord Bescande00 vloerrk er s 00
Figure 7: Upper vie,s' of the test set-up.
Figure 8: General view of actual test set-up.
7cD- V Actuator loads r- - LS
z '5 ---
5 - - - 1
1 0 5 IO IS 20 25 30 35 lime [mml - axial port - - -. starboard
Figure 9: Recorded actuator load histories.
During thetest n number of parameters were continuously recorded, such as the forces and displacements of the hydraulic actuators, and the axial and transverse displacement components of the micro-tang at port relative lo the mast section. Acoustic activity emitted by the mast was equally captured. as is well known, the acoustic emission technique is frequently used to monitor noise emitted by a siructural element to give an indication of the onset and growth of internal damage and fracture. The signals from the ultrasonic sensor were conditioned and amplified, and next transformed to audible frequencies so that the ucoustic activity could be easily followedduring the tests, hie recorded acoustic emission activity is shown in Figure 10.
Acoustic emission activity
A ,M 23' 35 40 linie [minj - Fjiort - acoustic emission activity
Figure 10: Recorded acoustic emitsion activity. The bearing strength used by the designer was in the order of 190 kN. and was in accordance with the Germanischer Lloyd Guidelines for example. The destructive lests however revealed actual bearing strengths ranging from about 63 over 72 (for the mast section without additional patching) to 88 kN (for the originalD2"-stage), well below the design strength. The Figure 1 shows the damage observed after the test, at port side. It clearly indicates the nature of the failureprocess, which is tearing of the micro-tang bushes through the laminate. Aller mast dissection de-bonding of the joint between both mast halves was found over a large distance.
Figure 1: Visible damage after test at port side.
Discussion Although the original mast design may have been in accordance to authority regulations principles indeed, it was estimated that reality was worse because the design and the fabrication did not comply fully with "good composite practice", and/or the consultation between designer and manufacturermay not have been done with the necessaiy care. The following remarks could be made concerning the "Pollux" mast design and fabrication:
Measurements whether by a company or by a research institute- of bearing strengths of pin and hole connections are generally obtained on high quality laminates which are flat or are at worst slightly curved. However, as shown by Figure 6, the mast consisted of two halves that were bonded. Bonded joints are always critical zones by themselves, with generally high three-dimensional stresses. It was therefore not a good idea of positioning the highly loaded micro-tangs precisely in the overlaps (or vice-versa)Figure 2). Especially with repeated loading, this may gie tise tuaridelaitunations over large disiances, and hence to disastrous buckling of the pre-bendeílniast. Figure 2: Micro-tang boreh oies were positioned in bonded overlaps. After the test a delamination was observed of the whole overlap over a large distance.
The design generally puts forward that bonds are of good quality with the bond joint laps well filled with resin adhesive. However, non-bonded arcas of typically 5-20 mm length md/or width were found to be spread over the bonded joint (Figure 13). Depending on the fabrication and bonding process, small de-bonds cannot always be totally eliminated. Designer and manufacturer have to consult on that aspect, and have to take this practical aspect into account in all phases,
Figure 3: Non-bonded areas of typically 5-20 min length and/or width ;s'ere regularlyfountL
High -too high- curvatures of the laminate were observed at the bond overlaps, which increase unwanted out-ofplane stress components, even with in plane loading (Figure 4). Moreover, high curvatures of relanvely thick shells make laminate compaction during the curing process difficult, which inevitably leads to badly bonded or even non-bonded interlaniinar zones. The section of the Figure 5. taken along the highly curved zone, demonstrates this. Non-bonded zones act more or less like delaminations, which especially under repeatedlfatigue loading- will gradually grow in these generally highly stresses overlaps. Figure 4: Sharp ¡amin ate curving at overlaps.
Figure 5: Badly bonded/non-bonded interlaminar zones in basic laminate along sharp1i' curved overlaps.
The additional diagonal patching at the "D"-stages seemed to be non-adequate. A mast not only has to take local loads, but equally global pre-stressing, associated pre-bending and eventually torsional loads. Boreholes (which are of course inevitable) disrupt the load bearing fibres. and are moreover stress raisers by themselves (typically by a factor of three to five). Correct load transfer asks for adequate extra patches, which by preference do contain non- disrupted fibres. The relative small iiifference between the measured static bearing capacities of the original D2-stage and that of the non-patched boreholes of the other tested mast section suggest that patching was not adequate: indeed diagonal patching of the Dl-stage was good for only 20-25% of the total load bearing thickness.
As a result of these remarks it is not surprising that actually measured bearing strengths were well below the design value that was in the mind of the designer. lt is generally accepted that due to the nature of laminated fibre-reinforced composites -of a non-homogenous. anisotropic and layered material, with the possibility of small internal defects and practically speaking no plasticity-, safe- operating loads under repeatedlfatigue loading conditions should stay well under 25%-33% of the static strength. depending on the loading spectrum and on the total number of cycles. That this is realistic indeed, could be observed also in the current static tests from the acoustic emission activity: more or less important internal damaging already was observed at about haLfway testing (seeFigure IO). Normal working conditions should thus remain well below, as once severe internal damage has occurred, it will tend to spread further under repeated loading. With the obtained static value of 88 kN a safe bearing load would thus rather be in the order of 22 to 29 kN.
As already mentioned. in order to verify real micro-tang loads against the design values, load measurements were performed on the lower mast rigging alter mast replacement. The result of the mast pre-stressing experiment is depicted in the figure 22. It appears that the resulting forces in the diagonal rods, and hence on the micro-tang are in the order of 27-29 kN (with the tension applied on the backstay), and are in accordance with the values used in the experiments.
u - - t..?Sn t t Itiji. tiuUtiTT%..0U.?t %. .tl tLttJ.J3_. 12 tensio9 applied on backstay (19.6 104 I
t ...... ckstay released Vipoft o .0 'cd t...rb
Di port
2 aT5j'a.-
- O 5
time [mini VI port - - - -. Vi starboard - Dl port DI starboard
Figure 22: Pre-stressing forces in diagonal and vertical rods offirst Stage.
The Figure 23 gives the recorded time histories of the forces in the diagonal rods while sailing at sea. These are the total forces, including the mast pre-stressing. The effect of raising sails (at about 8 minutes), and tacking (at about 42 and 49 minutes) is very clear. The loads vary hetween 30 and 50 kN, which taking into account the above remarks- gives the explanation for the early failure of the 'Tollux" mast.
Acknowledgement The author would like to gratefully acknowledge Mr. A. Verstraeten, owner of S/Y Pollux. as well as Proctor Yacht Spars. provider of the mact rigging, for the permission of using the experimental results for this contribution. -sage i - diagonalport side 60 sge i - diagonal - starboard
o o 10 20 30 40 50 60 lime min)
Figure 6 Figure 23: Recorded Jòrces in diagonal rods offirst stage during sailing (forces include mast pre-stressing forces). HISWA PAPER 2000 All Electric Yacht - Electrifying or Terrifying?
Dr. Ir. Ubald Nienhuis MBA' Netherlands Institute for Maritime Research
Eur. Ing. W. Folkersma2 TNO Bouw! Center for Maritime Engineering
Ir. J. van Vugt3 TNO Bouw/ Center for Maritime Engineering
Summary:
A yacht can be regarded as a smaller, personalized version of a cruise vessel. And the preferred solution for cruise vessels nowadays is a more or less filly integrated electric power plant. So the question of electrification of yachts inevitably surfaces in the continuous search for innovative designs.
While considerable efforts have been devoted to investigating the pros and cons of diesel-electric plants for commercial ships, this applies much less so for yachts. However essentially the same aspects to evaluating the alternative systems apply. But their weighting is different, and the cost!benefit equation shifts importantly with size, operation and type of ship.
This paper addresses the potential of diesel-electric propulsion for yachts. lt presents a look at the cost and benefit of different installations, the impact on the ship design and the impact on the building process. Part of the analysis is based on a realistic reference situation that comprises a state-of-the-art yacht with a corresponding operational profile.Lessons learnt from commercial shipping are transferred to the domain of yachts.
Conclusions on the applicability of electric power plants for yachts will be drawn, along with present- day constraints and challenges.
Introduction
Proponents of diesel-electric propulsion have presented compelling cases for the advantages of this concept. Advantages mentioned generally include: - Elimination of light load running of diesels, thus significantly reducing fouling of engines, - Superior emission performance, - Enhanced design flexibility allowing fairly arbitraiy placement of main power plant components, - Elimination of complicated shaft line arrangements, - Increased functional space (passenger, crew & cargo space), -Superior comfort in relation to noise and vibration or, alternatively, cheaper measures to combat these sources, - Higher system availability due to increased redundancy, - Shorter time-to-market due to the increased modularity of the system allowing more concurrent engineering and production,
Ubaki Nienhuis is an independent consultant to the marine industsy. Pail of his activities is on behalf on the Netherlands Institute for Maritime Research. For NIM he was project manager of the Dutch All Electric Ship project, which ran from 1996 fo 1999 and is continued in the form of the AES Platform with membership from 45 leading Dutch organizations. He is a graduate from De/fi Unwersity of Technology, Department of Marine Engineering. 2 Wiger Folkersma iscurrently employed by TNO Centre for Maritime Engineering (TNO-CMC), Dell?, the Netherlands. as Senior Scientist. As such he is amongst others involved in the analysis of power plant systems. Previously he was staff member technical and scientific computations at Vero/me Shipyard Heusden responsible for all strength, noise & vibration aspects. Hans van Vugt has been active in the design and development of control and monitoring systems for manipulators, robots and offshore platforms. He currently works as Research Engineer for ThJO Center of Maritime Engineering in The Netherlands. Since 1993 he is the pnme developer of the power p/ant simulation programme GES, working at TNO.
- -Quicker response of the ship due to quicker response of electric motors and constant speed operation of the diesels, - Superior controllability of the power plant, - Improved building logistics due to reduced dependency on long lead-time components, -Lower cost of installing the power plant, - Possibilities for reducing the amount of installed power, - Possibilities for standardizing the prime movers on board, thereby facilitating maintenance and reducing spare part inventory, -Relative ease of enhanced functionality, e.g. in adding dynamic positioning systems or introducing deeply submerged pumps, - Possibilities to eliminate gear boxes, -Possibilities to place the propulsion motor outboard, in the case of podded propulsion, -Improved potential for weight distribution to reduce internal stresses and thus reducing ship weight, -Reduced difficulties to manage shaft line vibrations (torsion, bending).
But opponents are quick to point out disadvantages, real or perceived, which make them hesitant to adopt it. They usually bring up arguments including: -Higher fuel consumption due to the additional conversions from mechanical to electrical power and vice versa, - Higher cost of the power plant components, Reduced reliability due to the increased number of components, -Higher demands on maintenance personnel both relative to their certificates as well as to their possibilities to effect repairs, -Demand on ship space to accommodate convertors, rectifiers and switchboards, -Increased difficulties for worldwide service & repair due to fewer repair & service centers for marine electrical equipment across the globe, Increased risk for builder due to lack of knowledge and experience with integrated electrical power plants, - Increased engineering time for systems with which owner and builder are less familiar, - Weight & volume penalties if gear boxes are eliminated due to the size of low-rpm electric motors, - Weight penalty in case of podded propulsor adoption, - Increased difficulties to manage harmonic distortions in the electrical network, - Acceptance by owners due to issues of image, perceived risk, etc.
All of these arguments are in principle correct. However they differ in importance depending on ship type, ship size and ship operation. Rational weighting of arguments is required, which must proceed on a case-by-case basis. But often, and unfortunately, the outcome is heavily influenced by less rational arguments which are based on lack of knowledge, lack of experience, lack of interest and lack of time.
The advantages are many. To throw away the advantages without attempting to overcome the disadvantages seems negligent to the present authors. Any marine stakeholder would do well to consider the issues carefully.
- Lessons-Learnt from Commercial Ships
The All-Electric Ship, or perhaps rather the Almost Electrical Ship, has made great inroads into certain niches of shipping. lt can be seen as the preferred solution for large cruise ships, for crane vessels, for certain types of offshore vessels and some other smaller categories. The majority of ships however (totalling some 96% of allinstalled power) remains (diesel)-mechanical. If AES is to become widespread, a cascade of events must be realised that together constitute a compelling case for AES adoption. Convincing owners must rest on a rational comparison of cost and benefit for all ships.
The Dutch national AES-project aimed at determining the costs and benefits for commercial ships typical for the Dutch shipbuilding and shipping scene. As such it forms an important step towards a more comprehensive effort on behalf of the Dutch industry to develop and introduce AES to ships for which diesel-mechanical plants still are predominant. Sponsors were the Royal Netherlands Navy; the Ministry of Economic Affairs; the Ministry of Transport, Public Works and Water Management; the Dutch Association of Contractors in Dredging and Shore and Bank Protection (VBKO); IHC Holland; Royal Scheide Group; Stork Product Engineering; Imtech Marine & Industry and HrvIA Power Systems. Some thirty parties, private and public, have been involved in the execution of the study.
Cost and benefit analyses require attention to issues such as emissions, maintenance, capital cost, hiel efficiency, reliability, availability and physical plant parameters. These aspects were converted to monetary values to allow realistic financial evaluation. However also more indirect consequences to ship design and hence to people or cargo-carrying capacity or building cost must be weighed in the final evaluation.
Several ships were analysed. These included a tug, a hopper dredger, a fish factory vessel, an inland dredging vessel, a chemical tanker, a frigate, a supply vessel and a cruise vessel. The investigations were carried out by teams comprising an owner, yard and plant equipment suppliers, aided by researchers and designers. In the course of their work widely varying plant designs were considered. In total some 50 different plants were analysed.
On the basis of such analyses challenges can be formulated. Figure 1 shows for some of the investigated cases (i.e. ship/power plant combinations and the year of realisation) the change in annual profitability due to AES when only taking into account component investment cost, weight, volume, fuel cost, maintenance cost and reliability. Note that negative values reflect an improvement. Required additional annual benefit [%] (cargo, comfort, building logistics, time-to-market) (Ship - Configuration/Year)
Harbour Tug - Diesel-Electric with 3 gensets/2000 Harbour Tug - Fuel cell-Electric/4 fuel cells/2005 Harbour Tug - Fuel cell-Electric/4 fuel cets/2020 iland Tanker - Diesel-Electric (4580 kW)/2000 leland Tanker - Fuel cell-Electric (21 000 kW)/2005 triand Tanker - Fuel cell-Electric (21 000 kW)/201 5 Frigate - AES II, 5 MW DG + 9 & 16 MW GT-SC/2000 Frigate - AES li, 5+1 019 MW DGsetsí2000 Engate - AES II, 5MW DG + 9 & 16 MWGT-KR/2005 Fngate-AES IV, 1Ox3MWFC/2015 Supply Vessel - Fuel cell-Electric (2348+22025 kW2O05 Supply Vessel - Fuel cell-Electric (2348+22025 kW2020 leland Dredger- Diesel-Electric with FP propellerl2000 kiland Dredger- Diesel-Electric with CP propelterl2000 Hopper Dredger - Diesel-Mechanic + Electric pumpsf2000 Hopper Dredger- Diesel-Electric and PP propellerl2000 Cruise Vessel - Diesel-Electric with CPP propellers/2000 Cruise Vessel - Turbo-Electric with FP pods/2000 Trawler - Diesel-E lectncI2000
-15% -10% -5% 0% 5% 10% 15%
Figure I - Economic impact oJelectrijicution
This figure thus also defines the extent of the challenges for AES to be economically attractive, which is here defined as a 'payback' period of less than 5 years. An example is the requirement of a 5% annual improvement in profitability for a DE inland tanker. These challenges are to be met by: improved building logistics and system integration; increased payload within the given main dimensions; increased comfort or saving on abatement measures; shorter time-to-market; and other benefits such as emissions and improved manoeuvring.
Also improvements in component technology, making them e.g. smaller, cheaper and lighter improves the balance for AES.
Some of the important conclusions of the rather comprehensive AES study for small and medium-sized commercial ships are: Already today, the advantages in revenue earning capacity enabled by having electric power plants clearly outweigh the current capital cost disadvantages for important classes of ships. In the near future the investment cost disadvantages of electric power plants is expected to reduce due to rapid developments in electromagnetic power technology. In the mid-term future fuel cells are expected to allow a lower cost of the power plant components, provided the car manufacturers live up to their expectations, and provided reformer technology on readily available logistic fuels are realized as anticipated. Electrification allows further standardization and modularization of power plants, which in turn yield important advantages in time-to-market, system integration & assembly cost and ultimately procurement cost, The difficulty in converting the clear 'intangible' benefits of AES (in e.g. noise, vibrations, emissions and controllability) into tangible,i.e. financial, consequences proves an important obstacle for the widespread acceptance of AES. Another obstacle for the adoption of AES is the inevitable learning curve and the consequent (perceived) risk in its application, which prevents most yards from a warmhearted embrace of this beneficial technology.
8o- 7. An essential ingredient in the chain of events leading to large-scale adoption of AES is the demonstration of its capabilities through designing, building, operating and, most importantly, monitoring, evaluating and sharing data of actual ships.
The Yacht Context
While the lessons learnt from commercial ships are valuable, they are by no means representative for the situation of yachts. Indeed the range of yachts is so wide that there can be no hard and fast rule relative to AES-adoption for yachts in general.
Let us try then to gauge the relative importance of factors impinging on the choice of power plant for yachts compared to the cruise vessel, for which integrated electric power plants are the preferred solution.
The first step in this assessment is comparing the issues that influence the decision: financial, intangible and otherwise. The below table typifies the most important cases by giving a mark on a IO- point scale for each of the issues. A score of 5 indicates that the issue is of the utmost beneficial importance. A score ofS indicates an important adverse influence. Note that the table only looks at the importance of the issue as affected by a change in the power plant. So really the table combines two questions: How important is the issue for the mentioned ship type in general, How much impact on the issue can be expected from introducing electrical power plants. A score of O for example can thus reflect that the issue is unimportant in general and/or that electric power plants have negligible impact on the issue. Note that the table reflects the situation of today, comparing diesel-Mechanical installations with diesel-electric. Turbine application and especially technological developments in power technology and hiel cells are expected to improve the situation for AES.
Issue Sail yacht Sail yacht Motor yachtMotor yachtMotor yachtMotor yachtReference (lmportance5Effect Small Large Small Small Large Large Cruise Ship of electric power plant) Slow Fast Slow Fast Benefit 0 8 4 7 14 15 24 Speed O O O O O O O Comfort 0 3 0 0 4 5 5
Emissions O I I 2 3 3 5
Range O O O O I I O Functional space 0 3 0 0 2 2 5 Stability O O O O O O O
Availability 0 2 I 1 2 2 3 Reliability -1 -1 -1 -1 -1 -1 -1 Maintainability -2 -2 -2 -2 -2 -2 0 Image 0 0 2 4 0 2 0
Controllability 3 2 3 3 I 1 3 Autonomy O 0 0 0 3 3
Time to market/client O O O O I I 3 Cost -6 -6 -8 -6 -3 0 -3
Fuel & lubrication cost O O -I .1 0 1 Personnel cost O O O O O O O Capital cost -4 -4 -5 -4 -3 -2 -5
Maintenance cost -2 -2 -2 -1 O I Other cost O O O O O O O
TTL APPRECIATION -6 2 -4 1 II 14 21
All of these issues are related to one another through the design of the ship. Naturally this table fails to capture all the richness inherent to all individual requirements and situations. Also it is merely a appraisal by the authors, no market research was performed. However it helps to focus on the right issues.
It is good to note that normally there is not a one-to-one relationship between the issue on the one hand, functional in nature and important to the owner, and the power plant characteristic on the other hand. technical in nature and important only to designer and builder.
Some of the mentioned issues merit additional attention: Maintainability. When considering an electric power plant as opposed to a conventional plimt. important questions in this respect are ao.: Accessibility ofthe equipment - Necessity for special personnel qualifications - Requirements for maintenance procedures and possibilities for condition monitoring - Modularity to ease replacing faulty parts or complete components With electric propulsionit becomes easier to modularize the power plant, allowing quick replacements of components. Of course this must be met by corresponding design measuresfor easy access to the engine room. s Availability. This is influenced by: -Inherent mean time between failure (MTBF) and mean time to repair (MYTR) of all'wired' components -Inherent rest capacity of faulty equipment, e.g. double-winded electromotors whichmaintain part oftheir output ifone ofthe windings fails -Inherent redundancy, i.e. spare capacity to make up for failures With electric propulsion, more components are normally introduced which by itself canintroduce a lower reliability. However simultaneouslyredundancy can be increased which leads to higher system availability. An example of this redundancy isthe fact that also the auxiliary diesels can contribute to propulsion power and vice versa. Further, one must be aware of thebenefits of inherent redundancy or gradual degradation of components themselves, as mentionedabove. Stability. For sailing yachts this is especially important. It would be veryadvantageous if the machinery could be placed lower down the hull. Unfortunately the size of the powerplant components prohibits this. Thus AES has no appreciableimpact on stability. Autonomy. For commercial ships this means for example beingindependent of pilotage, tug assistance or anchor-handling assistance. For yachts visiting pristine marineenvironments, restrictions to anchoring are increasingly imposed. Dynamic positioningis a suitable alternative. electric power plants facilitate this functionality. Controllability. For yachts a rapid response to changing situations, especiallywhile manoeuvring, is of great use. The response characteristics of propellers driven byelectric motors are advantageous compared to direct diesel driven propellers. This is one reason whycontrollability is improved in AES. Also the precision with which the running of the system can beadjusted is improved by AES. Fuel cost. For yachts fuel cost is rarely an issue. But fuel cost is affected by AES.On the one hand the overall powering performance is improved when the shaft line with brackets andbossings for twin-screw propellers can be eliminated. This requires the use of thrusters orpods which in turn allow a more optimum hull shape with consequent powering advantages.Also the improved operating conditions for the main engines lead to some reduction in fuel consumption. Onthe other hand the additional conversions from mechanical to electrical power and vice versalead to some loss of overall efficiency. In summary AES can be superior in fuel consumptionin certain cases. Normally however, it will be slightly less efficient. Speed. Speed, range and fuel consumption are related. Since changes inweight, overall energy efficiency and installed power are relatively small, speed is only marginally affectedby choosing the AES-concept.
The cost situation merits some additional attention. Yacht owners are notfaced with the same costírevenue equation as cruise operators. Still cost is an important aspect, if notfor the owner, then at least for the yard. For a cruise operator the capital cost is approximately 35%of his annual cost. The complete power plant installation amounts to roughly 20% of building cost. Thusthe power plant capital cost amounts to some 7% of annual cost. A 10% more expensiveinstallation, e.g. due to AES adoption, increases the annual cost by a mere 0.7%! The operator must recoupthese additional expenses by: - increased revenue from more passenger space; - by charging higher cruise fares on account of increased comfort; - by attracting more passengers due to a cleaner image and the permission tosail in pristine areas; - by profiting from reduced maintenance and fuel expenditure (if any). The cruise industry amply provides proof that this is easily feasible.
Now consider a yacht owner. Whatever way he looks at it, he isfuced with a similar cost equation, where capital cost also equals roughly 35% of annual cost. However his/hership is relatively more
82 luxurious and the power plant may consume perhaps 10 to 15% of the total investment cost. A 10% increase in the cost of the power plant, now raises annual cost by only 0.35%! But this must be offset by increased 'revenues', even if they are only intangible. Additional passenger space may not be all that important, since space is no longer at a premium for someone who wishes to own a large yacht. But it is at a premium for midsize yachts. But all the other factors are there for yachts, just as they are for cruise liners. Enough reason to suspect that also yacht owners may profit from AES or 'partly electric ships'.
Evaluation Tools
Rational evaluation of alternative power plants requires a cost and benefit analysis. Several factors influence the outcome of the equation, each with its own weighting. Factors are e.g. investment cost, reliability, maintenance, emissions, fiel consumption, weight, volume, noise levels etc. To this purpose the program General Energy Systems (GES) has been developed. It allows the user to compare arbitrary installations on a financial basis taking into account just such factors.
A new modelling language for domain independent physical systems is used. GES is based on object- oriented modelling to facilitate re-use of models and model parts. With the integrated graphical interface the user can easily build, modify and analyse the power plant. The components of GES are based on block-diagram structures, but the description of the components is textual. This combination allows efficient and elegant assembly of complex plants and the underlying comprehensive model libraries.
The philosophy of the program GES is that developers specialised in different fields can independently model their components. GES allows easy and self-correcting coupling of these different components. This is done by describing the system boundaries of each component model with a pair of variables that describe the exchange of energy with another component.
For every physical domain a pair of variables is available to describe the energy flow, for example number of revolutions and torque.
For every ship an operational profile is determined. The operational profile describes the different power consumers and their consumption over time. To this end the relevant period is separated into an arbitrary number of time intervals. For every interval the stationary energy equilibrium is determined, taking into account all characteristics of all components. j I . Là '. -. L.a L
aI 2
Ç!fl Ç!
Ç!- a Ç! - a !- -- 1 [-- G.ae,-- I -a D - Ç! Ls a Sa Ç!a Initially the fixed attributes of all components are determined such as mass, volume and cost. These are then summed over the installation. Subsequently load-dependent attributes such as fuel consumption, reliability, emissions and availability are calculated. All output is then written into a spreadsheet file that allows the user personalised access. An EXCEL macro is available for graphical representation of the results and financial evaluation.
For the All-Electric Ship study a large component library was developed within the GES framework. The library consists of all major equipment for ship installations. A unique feature is the combination of physical and economic parameters. Several principles can be used to store the data. Both generalised formulas as well as tabulated data lists can be employed. The components are grouped according to their main energy domain in a tree structure. With simple drag and drop a required component is copied into the model. The equations of each component model are written in include files to allow the user to override values based on experience or data. Instantiation of components ensures that components need to be specified in the form of algorithms only once. Figure2shows a sample installation based on future application of fuel cell technology.
Cost & Benefit for Yachts
As an example we have studied a75m displacement yacht, fitted with twin fixed-pitch propellers, capable of speeds upto 18 knots. The reference power plant consists of two main diesels of1850kW running at 1000 rpm connected to the propellers by gearboxes. Next to that three diesel generator sets of260kW each provide auxiliary power for air conditioning, domestic loads, transverse thrusters etc.
As an alternative an integrated electric power plant concept was studied comprising of three main generating sets(1275kW each) as well as three auxiliary sets of260kW each. The propellers are driven by1850kW direct drive electric motors. Both configurations are shown in Figures 3 and 4.
flE .rhj E Lea jB : U*
&d h_MDU
W..jen O-e..e,
r,e
e ¡ - -- ej
-
q,7 IA_Dek Powe, 130w
Figure 3 - Diesel-mechanical configuration 11FRIY.onhll NUX f' Lme,L'rs â 1n e dow ..I2J
Figure 4 - Diesel-electric configuration
Two hypothetical operational profiles were used which describe: the yacht being sparsely used by the owner. the yacht being heavily used by owner or charterer.
Figure 5 shows the operational profile for the first situation that comprises yachting at cruise speeds both in the Caribean and the Mediterranean. with fast transit journeys in between, and much time spent in harbours drawing on own energy supply.
Ti9Yunhtntnhuunel, Techno'ogy 1998 Runde30-09-2000112.4522 Input finOPyecht_I cuy Operational Profile, Required Power [kW Outputten:Yucpt_MECH_I ceo Renu,Ilu fltnReeult_MECH_1 cuy 3600
3000 Thrusters 2600 D Starboard_Propeller Port_Propeller
2000 LightIng Alr_Condltioning Accom m odatlon i600 D Galley Safety_Systems
Engine_Service
600
o 11 o o ¿ ¿ ¿ C g o fe C . - - - Ç, Ç- t n n t L) Z Z 13 J u o - r 0_ iY, 222'-2 CZ w C'- 32 5 ' 0 w '- OnOn Qn== c o o 5 1gW.- 1g0.- 1g = C 0- 9 w Q = = C O C t: t: t: t: c n ut ut w = 2 2 2 o o V k- L) O LI L)
as For the calculationitis necessary to input not only theoperational profile and the power plant configuration, but also the resistance characteristic of thevessel. The GES calculations then yield investment cost estimates per component along withdetailed projections of maintenance cost, fuel cost, availability, emissions and weight and volume of the powerplant components. A sample result shows the emission estimates for the diesel-mechanical situationand the above operational profile, Figure 6. Figure 7 shows the annual fuel consumption for one ofthe diesel-electrical installations.
T8eYsopj n,edesoc. Teshootol998 Runde3O-O9-2OOOIlZ.4522 EMISSION (ton/year] Input ñlsOPystht_tcsu Output file Ysnht_MECH_t esu Results ñlsResuIt_MEÇH_tesn 14
12 600
10 600 C USum app. of CO2 (tottlyearl 8 USum app. of SOx [toolyearj 400 O USum app. of NOs (tonlyear] C O o OSum app. of NC ton1year1 nl 6 o300 O Sum app. of CO tonJyaar1 o z 4 200 u)
2 10:
o C C C s t, C C t W O t (3 C C (3 a C C C -C C o o n o C C 2 2 2 e C 2 2 2 e X Q O O O Q Q o X n C C C ot = C C a. n O C C o n n n s C -c -c O O o O O
Figure 6 - Annual emissions for diesel-mechanical installation
TiOe.Yeslit AES. T1dcey 1998 Run dateOl-1O-20«1i16:1050 FUEL CONSUMPTION (ton/year] Irtpstt tleOP_ya39_1_C.Csv Output 9eYnt_AES_1_c.csv Results file Result_AES_1_c cnn 250
200
aux_diesel_unit_l, tank_2 aux_diesel_unit_2, tank_2 150 aux_diesel_unjt_3, tank_2 Ddiesel_unit_1, tank_2 Udiesel_unjt_2. tank 2 100 dieselunit_3, tank_2
50
F14 u f 14.1 IT'C Q o, C C C s C Q C- tC C'I t 'W Q C a, et a, m a Q Q 0 = o C o Q a n o C o t E E E 2 o Q z t.? O O C t o = G t t o C Q a. Q Q Q Q Q n o n n n n t Q Q C o Q Q Q Q O I- The table below shows a comparison of theinstallations for the four considered cases.
Eleril/Heavily Used Mhanical/Sparse Used Methanical/Heavily Used Dectncal/Sparsely Used 100 121 121 Investmentcost[kf lOO 100 147 147 Compnentvcume(m' 100 115 115 100 100 ComJnentmass1tonl 90 96 Maintenancost[kfiyrl 100 125 100 100 100 Availability[%J 100 154 107 172 Fuel cost [kf/yr] 100 155 107 172 CO2[tcyVyr] 100 154 107 172 SOx[tcilyrJ 100 178 86 142 NOx(ton/yrl 100 137 89 166 HCtorVyrJ 100 146 82 157 CO[tor'yrl 100
It can be seen that: for other ships The electrical installation is moreexpensive. This result is in line with that found - Note that also and is largely caused by the higherexpenditure involved in converters and rectifiers. the cost shown here includes component costonly. Cost benefits due to improvedbuilding logistics and easier installation are not calculatedby GES. Similarly it - The mechanical installation is smaller whencalculating the added component volume. is somewhat lighter. This is caused on the onehand by the switchboards and associatedconverter motors which and rectifier equipment, on the otherhand by the rather heavy direct drive electric eliminate the need for gearboxes. Also here notethat only component volumes are calculated: no account is taken of the possibilities toimprove accessibility and geography. prime The electrical variant consumes slightly morefuel because of the selection of smaller - because of the additional movers with a higher specificfuel consumption at nominal load, and energy conversions. This is notquite offset by the higher loadingof the prime movers which contribute to better fuel consumption. -The electrical variant is superior in termsof maintenance and noxious emissionsdue to better loading of engines. facts cruise Note that this result is quite similar to thatfound for cruise liners. Indeed despite these benefits. operators overwhelmingly adoptelectrical installations on account of all other consists of 3 main These results refer to the initial design casein which the diesel-electrical installation of the loading and 3 auxiliary diesels. GES clearly showsthat this installation is not optimal on account installation can of the prime movers. Distinct improvementscompared to the presented diesel-electrical column) shows the be found by changing the choice ofthe prime movers. The below table (second estimated effects of such improvements,maintaining all other parameters as before.
10years Mechanical/Sparsely Used Electrical Optimized Electrical in 93 100 114 Investment cost [kfl 100 100 140 Component volume [mA3J 110 100 110 Component mass [tonj 80 Maintenance cost [kf/yr] 100 80 100 100 100 Availability f%l 98 100 107 Fuel cost [ldlyr] 98 100 107 CO2[ton/yr] 98 100 107 SOx[ton/yr] 64 100 70 NOxIton/yr) 73 100 80 HC[tan/VIl 69 100 75 GOtoniy secondary emissions. Improvements are obtained in investment cost,volume, weight, maintenance and developments. Mass fabrication of powertechnology and the The AES study has also looked at future here. The last introduction of much cheaper pods all have animportant bearing on the issues presented Note that all other column of the above table shows an estimatewhat may happen over the next years. successful in benefits of electrical propulsion remainin this case as before. If the car makers prove their developments, the introduction offuel cells may lead to significantfurther advantages. Opportunities & Challenges
For commercial ships, a picture emanates from the AESstudy of the critical issues that affect the widespread acceptance and hence adoption of AES. For yachtsthe situation is different but important similarities arise. The challenges then can be summarized as follows.
General For an efficient and fast comparison of alternative power plants,taking into account operational profile, layout of the vessel, investment and operational costs, a toolsuch as GES is of prime importance and must be made accessible for the yacht industry.
Ship design The present investigations into the application of electric powersupply in ship types, which are traditionally provided with a mechanical propulsion system,confirm that for a fair comparison between different alternatives, the complete system has to be considered.
A simple replacement of the mechanical propulsion systemby an electric system will in all cases result in an increase of investment and in most cases also anincrease in the! consumption. Consequently such a (quick and dirty) comparisonwill almost certainly lead to the conclusion that AES is notthe way to go.
When, on the other hand, the complete system is considered,the choice of an electric power system is much more likely. It allows the designer to includethe operational profile of the vessel, to optimise the power management on board and tooptimise the layout of the interior. Further he can make useof positive effects which an AES concept may have onthe reduction of noise & vibration abatement measures, comfort of passengers and crew, onthe reliability of the power supply, on the nature and quantity of exhaust emissions, on the fouling and maintenance ofprime movers, etc.
In this area lies one of the main challenges for AESadoption for yachts. The designers must rise to this challenge.
Electromagnetic equipment Generators. In the short term the available low speed generators cansupport the All-Electric Ship developments. Future developments will have to focus onreduction in size and weight of the generators. The already on-going developmentsof high speed and permanent magnet generators allow for this reduction. Motors. Developments in motors will focus on the efficiency,improved heat transfer and the associated dimensions of the motor. As DC motors have a serious maintenancedrawback compared to AC motors, they will rapidly reduce in number. Promisingdevelopments are permanent magnets that allow for reduced size of the rotor. Developments both in axial andradial flux motors will have the further advantage of a compact form in applications where space andweight are at a premium. Converters. Reduction of size is to be expected fromdevelopments in semiconductors and power electronics. Important cost reductions are foreseen in viewof the widespread application in land-based energy distribution. Accurately switchingthyristors will in turn require fewer control-electroniccircuits and the harmonic distortion will further be improved.Moreover they make multi-voltage levels possible and will allow for increased power at high voltage.This in turn will reduce the current-related size of wiring and switching equipment. Transformers. As power increases the use of mediumvoltage primary power systems will be common. Nevertheless low voltage consumers will remain. Thisrequires MV/LV (MV: medium voltage, often 6.6 kVolt but also e.g. 690 Volt; LV: low voltage, often220/380 or 240/440 Volt) transformers Lo be installed. Again, size and weight are aspectsthat should be addressed in order to reduce the implications on the ship design.
Prime movers -Most conventional prime movers (diesels, gasturbines) for maritime use are built for direct propulsion, with a very large variation in load and rotationalspeed. Prime movers for electrical generation however, perform within narrow limits and possibly evenat a fixed rotational speed. Optimisation of these prime movers for maritime use couldwell offer advantages in efficiency, maintainability, reliability and environmental issues. -8- A real challenge is the development of a fuel cell system,that can compete with a diesel (gas turbine) generating set in dimensions, weight, efficiency and especially cost.Being scaleable, they offer further important advantagesinship design, maintenance and operation, along withquietrunning characteristics. In most cases these fuel cell systems shall have tobe operated from logistic (diesel) fuels. This implies developments in the reforming process, asthe fuel cell inherently needs (more or less pure) hydrogen and oxygen. If the car makers live up totheir expectations fuel cells may be expected to enter the yacht market in the near future.
Propulsors & other consumers The podded propulsors have become the preferred solutionin cruise ships and applications in offshore are increasing. Submersed electric motorshave stimulated precision deep dredging and trenching, creating new markets. State-of-the-art electric motors inthese applications will have to be maintenance free, with smallest dimensions and effective cooling.Oil-filled AC cage induction motors driven by pulse width modulation converters (PWM) are the projectedingredients also for the upper power range. However the weight of the propulsors form an importantobstacle to its widespread adoption for other ship types. Developments inelectric motors can only partially remove this problem,and the conventional, electric motor driven thruster will deserve carefulattention from yacht designers.
Multistage propulsion with twin co-rotating or contra-rotatingpropellers offers a perfect application for electric propulsion. In podded systems the rotor and 'stator'could run at different speeds, each driving a propeller.
Shaft-less propulsion through magnetic bearings andsuper-conducting electromagnetic thrusters are interesting, yet many of their features require more research.
Installation & integration Most long term All-Electric Ship configurationsinclude fuel cells as the most promising power generation system. Application of fuel cells meansapplication of DC switching equipment and protection systems. The installed power will be in the rangeof MW and tens of MW, again forcing the system voltage level to medium DC voltage.The availability of MV DC switching equipment is a prerequisite for the implementation of any MV DC power system.The low voltage consumers, conventionally AC, or in the future DC will need conversionequipment to connect to the primary MV DC power system.
The All-Electric Ship will break the traditional waysof shipbuilding and ship operation. Plug-in components profoundly affect the logistics ofshipbuilding with the possibility of increased modular approach to shipbuilding and re-engineering the assembly process.This defines challenges to the system integrators to simplify connectionand powering-up processes, and achieve 'plug & play'.Also standardised sub-systems with advantages in cost andtime-to-market will be greatly stimulated by AES.
Operation Plug-in components also challenge the maintenance atthe original equipment manufacturers (OEM) versus onboard repairs. Increasedship availability will result while lifetime andperformance of equipment and overall safety in a potentially hazardousenvironment is enhanced. Service will become more and more a land-based activityrequiring an extensive logistic support.
Operating an All-Electric Ship will require skills in thefield of power conversion equipment and especially training in use and maintenance of MVequipment. Alternatively comprehensive service contracts will have to cover also the electricalinstallation. Development ot safety standards and procedures, common for land based installations, dedicated toships will have to support crews and owners in safely operating the ship.
The All-Electric Ship concept will make extensive useof computer based control systems. Embedded controllers, networks, interfaces and workstations arebecoming new area's in which future technical staff need to be educated in order to successfully operateand maintain an All-Electric Ship. Here there is another challenge: to be able to use monitoring systems tooptimise (preventive) maintenance, to assist in running engines for minimum emissionsalong with other functional support to owner and crew.
- Why Should Yacht Builders Bother
The figure below shows the market situationfor yacht construction as of 1996. lt confirmsthe dominating position of the three main yacht-buildingnations Italy, US, and Netherlands, the latter having the largest share in the large yacht business.Innovation is essential to maintaining this position. Introducing AES to the yacht business may help toconsolidate and expand this position, especially in view of the potential advantages of the AES concept. be In view of the previous discussion AES seems tooffer most benefits for large yachts. Hence it may AES. If expected that the Dutch industry stand most togain (and lose) from the developments towards where designers, suppliers one further considers the strengthof the maritime cluster in the Netherlands would do and builders cooperate intensively, it can beconcluded that the Dutch yacht building sector well to pursue further investigations into AES.
300
250 a, 200 I 150 I
100 2 E z 50 1 o t
tO Q) O) q - u O O N (j) z Country of build
B No of ships 30-36 m No of ships 36-45 m DN0 of ships over45m
Figure 6 - Market distribution yacht building (1996)
Concluding Remarks
The concept of AES is no doubt here to stayfor a long time. But for it to mature and becomethe preferred solution for yachts requires the concertedaction of all stakeholders. The success of AES lies not in isolated new technologies, howeverimportant that contribution is. Rather it lies in: exploiting the advantageous synergies that electricalgeneration, distribution and consumption offer when properly integrated into the systemof the ship. Examples are innovative solutions to ship design resulting in increased passenger space; improved building logistics and hence lower totalcapital cost; and further standardisation with resultant lower component and integration cost; improved comfort levels; with better operational integration yieldingimproved availability, rcliahility and maintainability the consequent advantages in cost and safety; increased functionality through the addition ofdynamic positioning; access to fragile environmentsand improved sustainability through loweremissions; enhanced image of the yacht exploitingcutting-edge technology. of The Dutch maritime industry continues to co-operatein pursuit of these possibilities for all types that ships; to reap the benefits that All(most)-ElectricShips offer for end-user, industry and society. To have decided to end over 40 leading organisations activein shipping, shipbuilding and its supply chain join the national AES-Platform(www.aesplatform.nl).
9Ó WIND TUNNEL TESTS ON SCHOONER RIGS AND THEIR USE IN PERFORMANCE PREDICTION BY VPP CALCULATIONS
by I.M.C.Campbell and G. Dijkstra
HISTORICAL BACKGROUND
A tour around the maritime museums of the world will reveal the rich history of Schooner rigs used to propel commercial sailing vessels. Just a few examples are; the Schooners built in Dartmouth in the UK to ply the coal and fruit trade to the Mediterranean; those that sailed out of Paimpol in Brittany to fish off Iceland; and others that sailed from East coast ports in the USA to fish off the Grand Banks. It was clearly a universally successful working rig that also found favour in cruising and racing yachts, with America winninga race around the Isle of Wight almost 150 years ago to take the "America's Cup" to the New York Yacht Club.
In recent years a number of devotees have kept the Schooner rig alive and well through restoration, re-build and new-build projects. This paper describes tests conducted with wind tunnel models with the aim of continuing the development of the Schooner rig.
TECHNICAL BACKGROUND
It can be difficult to fit all the required sail area on to a single mast for a large yacht. Small sloop-rigged yachts can be scaled up to larger sizes but to have the same scale performance, according to Froude scaling, the wind speed should increase with the square root of scale. This is less likely to occur in practice, since for example a large and small yacht sailing together operate in the same wind speed, except for the increase with mast height due to the wind gradient. It can therefore be desirable to increase the sail area of a large yacht relative to that of a small yacht.
Two-masted Schooners and Ketches provide a means to increase sail area by giving an overall trapezoidal shape to the rig compared to the triangular shape of a Sloop. Both the overall and centre of effort heights of a trapezoidal rig will also be lower than that of a triangular rig of the same area. A lower overall rig height will reduce the aspect ratio of the rig withan associated reduction in the aerodynamic efficiency of the rig through the increased induced drag due to lift. The lower centre of effort height will offset this, in part, with a consequential reduction in the heeling moment, which may give períòrmance advantages in some conditions.
There are a wide variety of Schooner sail plans involving combinations of sails set flying from or attached to the mast, which makes selection of the sail plan interesting.
9' RIG DESIGN PROBLEMS
The wind tunnel tests described in this paper involved the study of the two-masted classic Schooner rig shown in Figure 1. The features of the "classic rig" may be listed as follows:
A higher mainmast than foremast. A bowsprit, from which the outer jib is set. Two other similar sized headsails - an inner jib and staysail. A foresail, with a sail plan to fill most of the between mast area. A Bermudan mainsail set on a long boom.
Design problems for this type of schooner rig include:
Optimisation of the headsail arrangements within the fore-triangle to avoid interference problems between the sails and to maximise their performance. Optimisation of the foresail, with sail options being a gaff foresail with topsail, a wishbone topsail or a modem fully battened large roach foresail. Design of off-wind sails to be flown between the masts, without undue interference with the foremast. Consideration of reefed sail configurations to maintain balance with the hull.
The "modern" Schooner rig, which is shown in Figure 2, was tested to investigate some of these design problems.
WIND TUNNEL TEST ARRANGEMENTS
The Wolfson Unit conducted the tests in the wind tunnel at the University Southampton. The low-speed test section is 4.6 metre wide by 3.7 metre high. A complete model consisting of hull, mast rigging and sails was mounted on a six-component dynamometer, which is fitted under the floor of the wind tunnel.
The wind tunnel arrangement enables the sail forces to be measured without any reference to the performance of the yacht, whereas when sailing the sail forces cannot be measured directly and the effect of any changes can only be assessed from changes in the velocity of the boat. The wind tunnel is operated at a constant speed, thus eliminating the effect of fluctuations in wind speed that make data gathering from full scale trials difficult. The requirement to trim sails to suit different wind speeds has to be simulated by the experimenter since the rig is firmly attached to the dynamometer and does not alter in heel during a set of runs.
Further details of the experimental setup are described in reference 2.
THE SCHOONER RIGS TESTED
To aid the development of Schooner rigs for the design of particular yachts a number of wind tunnel tests have been conducted at the University of Southampton. The data presented in this paper was derived from tests for two yachts of slightly different size but sufficiently similar to be represented by the same model. The data have been scaled from the model test results by a common factor of 1:25 and the size difference between the yachts' rigs was treated subsequently by applying additional scale factors. 92- The sail plans are shown in photographs with examples of the full up-wind and off-wind sails used in the tests shown in Figure 1 for the classic rig and Figure 2 for the modern rig. Because the tests were related to different designs, they were conducted in different test sessions spaced several months apart, with the model refurbished and reinstalled into the wind tunnel for the tests on the modern rig.
The classic upwind sail configuration consisted of three headsails, a gaff foresail with fore topsail and a Bermudan mainsail, with a total sail area of 1225 m2. The modern upwind sail configuration consisted of two headsails, a fully battened large roach foresail anda fully battened mainsail, with a total sail area of 1146 m2. The smaller sail area of the modern rig was principally due to differences in the mainsail area and headsail areas. The height of the main mast above the datum water line was 47.9 m for both the classic and modern rigs.
The modern rig is designed to produce improved performance by taking advantage of modern sail materials and sail handling systems.
The classic, or nearly classic, off-wind sail configuration consisted of the upwind rig with the outer jib replaced by an asymmetric spinnaker and the foretopsail replaced by a fisherman sail, with a total sail area of 2046 m2. The modern off-wind sail configuration consisted of the yankee jib replaced by an asymmetric spinnaker and the foresail replaced by an asymmetric main gennaker, with a total sail area of 2314 m2.
The modern off-wind rig is designed to produce improved performance by taking advantage of modern asymmetric sail shapes to increase the sail area.
SAIL SHEETING CONSIDERATIONS
The model tests in the wind tunnel provide the opportunity to study the sheet leads, for positioning deck fittings, and sail interactions, between themselves and with the standing and running rigging. The three headsails in the classic rig overlapped so variations in relative tack positions and sizes were tested to determine combinations that avoided interference between the sails. The foresail is an interesting sail to set because it is between the headsails and the mainsail and there are various combinations of boom angle and sail twist that can be used. It was found that to some extent the foresail sheeting had rather less effect on the driving force and heeling moment than the sheet adjustments to the headsails and mainsail. Sheeting information such as this could be used to help the crew operate the yacht and the designer optimise the winch requirements.
The fisherman and the main gennaker can both interact with the foremast rigging, furthermore the gennaker should be set to avoid chafe with the foremast runners and the lower triatic stay. The interaction depended on the sail sheeting and its optimisation at different apparent wind angles. At apparent wind angles of greater than 90 degrees the interactions with the rigging restricted the extent to which the gennaker sheet could be eased and in some conditions this limited the maximum driving force that could be achieved. At apparent wind angles of greater than 135 degrees significant blanketing effects occurred between the sails, which would restrict performance and the extent to which the yacht could be sailed downwind.
All the data points measured, including those presented in this paper, followed trimming of the sails to optimise the driving force for the required heeling moment. Initially, the
-9__ maximum driving force was sought irrespective of the heeling moment, then the heeling moment was reduced in stages and the highest driving sought for each condition.
DATA FROM TESTS
The test data presented in this paper are for particular up-wind and off-wind sail combinations for the two Schooner rigs. They are representative of much larger sets of data for various sail combinations that were tested to aid the rig development for the yachts.
Corrections were applied to the data for zero drift during a set of runs, which could typically involve 15 minutes between starting and stopping the wind. Wall boundary and wake blockage corrections were also applied as described in references 2 and 3. The forces and moments were measured on the body axes of the model, which was set a zero leeway angle on the dynamometer. These driving and heeling forces were transformed in the analysis to the apparent wind axis and normalised by the sail area and dynamic wind pressure to produce the lift and drag coefficients.
AERODYNAMIC CHARACTERISTICS REVEALLED BY THE TESTS
Driving forces and heeling moments
The driving forces and heeling moments affect directly the performance of a yacht. When the heeling moment is in balance with the righting moment the driving force, corrected for heel angle, must be in balance with the drag and this will determine the speed of the yacht. The Velocity Performance Program (VPP) iterates to the equilibrium of these aerodynamic forces and moments with the hydrodynamic forces and hydrostatic moments. It is, however, clear that if, at a particular apparent wind angle and heeling moment, the driving force from one combination of sails is greater than from another then the speed of the yacht will also be greater.
The curved lines shown on Figure 2 are calculated from values fitted to the lift coefficient, drag coefficients and centre of effort height, using the same algorithm as is in the VPP. It can be seen that a reasonable fit was achieved to all the test data across the full range of heeling moments. It should be borne in mind when studying the data in Figure 2 that the reduction in heeling moment at unit wind pressure represents those adjustments in the sheeting of the sails that would be made when sailing in increasing wind strengths to control the heel angle. The heeling moment may increase due to the increase in wind pressure.
It can be seen from the data that at apparent wind angles of 25 and 30 degrees the VPP fit and the data tend to a maximum value for the driving fOrce that would not increase with a further increase in heeling moment. The maximum performance of the rig at these angles is limited by its aspect ratio and the influence of this on the increase in induced drag due to lift. At wider apparent wind angles the induced drag has less effect on the driving force and the VPP fitted curves show an upward trend at the maximum heeling moment. The maximum driving force at apparent wind angles greater than 36 degrees is limited by the maximum lift that can be developed by the rig.
Comparison of the data from the tests on the two rigs indicate that at apparent wind angles of 25 and 30 degrees the modern rig produced significantly higher driving forces than the classic rig.Thus the modern rig would have better windward performance despite having a smaller _9L_ sail area than the classic rig. At an apparent wind angle of 60 degrees it can be seen that the classic rig produced higher driving forces than the modern rig, probably due to its larger sail area.
Windage
The total windage of the hull, deck, mast and rigging was measured by removing the sails from the model. The windage forces were subtracted from the total forces obtained with the sails to produce the residual sail coefficients. The windage forces were then transformed to lift and drag coefficients using the upwind sail area as the reference area. This enables straightforward comparison of the windage coefficients with the sail coefficients.
If wind tunnel test data is used in a direct manner to compare different sail then this should be done on the basis of the total sail forces or moments. If, however, the wind tunnel data is used to derive coefficients for use in Velocity Performance Predictions (VPPs) then the windage should be input as a separate component to the sail forces. This ensures that the VPP reefmg routine operates properly, otherwise the VPP will invoke the reef function at too low a wind speed because it will incorrectly reduce the windage area with the sail area.
Since the windage forces were subtracted from the total forces in the derivation of the sail coefficients it is necessary to use the same centre of effort for the windage as that derived from the heeling moment and heeling force for the rig with sails.
The windage lift and drag coefficients are shown in Figure 6 together with the shape functions used in the WinDesign VPP for the variation of windage with apparent wind angles. It can be seen that the shape functions do not match the test data over the full range of apparent wind angles. Since windage is most significant in affecting windward sailing performance the windage areas and coefficients were adjust in the VPP to match the driving forces to those from the tests over the apparent wind angle range of 25 to 45 degrees. This can be seen from Figure 5, which contains the measured windage forces.
Variation of drag with lift
When the sail forces are transformed from the body to wind axes the data from tests at different apparent wind angles tend to collapse as can be seen in Figure 3. This indicates that the aerodynamic characteristics of the rigs are determined by their overall shape, planform and aspect ratio, despite the changes in sheeting required as the apparent wind angle is increased and despite the associated changes in the relative positions of the sails.
Comparison of the data from the tests on the two rigs indicate that the modern rig produced significantly less drag at the higher values of lift than the classic rig.This is indicated by the dotted lines shown in Figure 3, which are values fitted to give an effective rig height, He. lt can be seen that this is higher for the modern rig.
In the Hazen aerodynamic model, reference 4, there are two components of drag that are proportional to the square of lift, the induced drag due to lift and a component of viscous drag which also varies with lift. This viscous component is a smaller part of the total drag. In reference 2 it was shown that the Hazen model could be fitted to wind tunnel data from tests on a Bermudan sloop rig. The data from the Schooner rig tests had, however, slightly different characteristics, with a tendency for a slight curve in the variation of drag coefficient
9 with the square of lift coefficient. This tendency could be attributed to the effect of twiston induced drag, as the sails were eased to reduce the heeling moment.
The data fit for the VPP was obtained by first fitting a curve through the variation of lift with apparent wind angle, as shown in Figure 4, and then obtaining the associated maximum drag coefficient from Figure 3. The effective rig height is used to obtain the reduction of drag with lift as the sails were eased. At zero lift, the intercept of the line associated with the effective rig height was not zero drag, despite having subtracted the windage forces from the data. This was partly due to the fit of the line to the curve of data, as described above. Caution must therefore be taken in the interpretation of the values of effective rig heights of 49m and 51m, for the classic and modem rigs respectively, compared to the geometric rig height of 47.9m.
Lift coefficients
The variation of lift coefficients with apparent wind angles for the upwind rigs is shown in Figure 4 together with a curve fitted through the values associated with the maximum driving force for input to the VPP. It can be seen that the modem rig tended to produce slightly higher values of lift coefficient than the classic rig.
The variation of lift coefficients with apparent wind angles for the off-wind rigs is shown in Figure 8. These show a maximum value at an apparent wind angle of 60 degrees, with a progressive reduction towards zero lift at an apparent wind angle of 150 degrees. At an apparent wind angle of 90 degrees all of the lift contributes to the driving force. It can be seen from the results of the \TPP calculations, given in Table 1, that this angle is only reached when sailing in the optimum downwind Vmg condition in a true wind speed of 14 knots.
Comparison of the lift coefficients for the classic and modem off-wind rigs shows that the modem rig produces higher values. This may be attributed to the increased camber of the main gennaker compared to that in the foresail and fisherman. The combination of increased lift coefficient and increased sail area of the modem off-wind rig significantly increase the driving force.
Drag coefficients
The variation of drag coefficients with apparent wind angles for the off-wind rigs is shown in Figure 9. At apparent wind angles of less than 90 degrees the drag acts to reduce the driving force so low values for a given value of lift are desirable.
Comparison of the drag coefficients for the classic and modem off-wind rigs shows somewhat different shaped data fits. The higher drag coefficients that the modem off-wind rig produces at an apparent wind angle of 75 degrees are associated with the higher lift coefficients. It is interesting to note that the drag coefficients from both rigs are similar at an apparent wind angle of 120 degrees, where the drag contributes to the driving force.
Centre of effort height variations
Figure 7 shows the variation of the centre of effort height with the heeling force coefficient (Cy) for the upwind rigs. Both the classic and modem rigs showed a clear trend of reducing height with reducing heeling force. This is attributable to two factors:
9 Firstly the heeling moment is strongly affected by the sheeting of the main sail, which has the highest centre of area of all the sails. As the main sheet is eased so its contribution to total aerodynamic force is reduced and the overall centre of effort height tends towards the lower centre of area of the sail plan of the foremast;
Secondly the twist in both the mainsail and the foresail increased as their sheets were eased to reduce the heeling moment. The twist could be controlled by the relative tension in twin sheets, led to either side of the centreline, but allowing twist was found to optimise the driving force at reduced values of heeling moment coefficients.
The WinDesign VPP has a linear reduction of centre of effort height with the flat function, which is used in the program to reduce the lift coefficient to control the heel angle. The factor in the VPP for the rate of reduction was determined from the test data.
Another trend that is particularly apparent in the data from the modem rig is the reduction in centre of effort height with apparent wind angle. In the version of WinDesign used in the VPP analysis for this paper there was no function to simulate this trend but will be incorporated in future versions. This increased the difficulty of the task of fitting values of lift and drag coefficients to give a match to the driving force and heeling moment data across the range of apparent wind angles from 25to 60 degrees.
VPP CALCULATIONS
Program and inputs
Performance calculations were made using the WinDesign VPP, which is described in reference 1. Sail coefficients are input to this program as a table of apparent wind angles with maximum lift coefficients and the associated drag coefficient, together with the reference sail area, centre of effort height and effective rig height. This height is used to obtain the reduction of drag with lift when the VPP applies the flattening function to control the heel angle of the yacht.
Data fits
Although the table of inputs to the VPP for sail coefficients appears simple, care must be taken to ensure that there are fair curves through the variation lift and drag coefficients with apparent wind angle, because the VPP will interpolate between the tabulated values. The input lift and drag coefficient must also represent the experimental driving forces and heeling moments from which they were derived.
The data fits were derived by plotting them on the charts in the spreadsheets that were used to analyse the data. The driving forces and heeling moments were recalculated, using the same algorithm that is in the VPP, from the fitted lift and drag coefficients and the selected centre of effort and effective rig heights. The process involves manual manipulation of the data to obtain the best fit over the full range of test conditions.
The results from the VPP calculations are given in Table 1. Up-wind performance
The VPP predictions for the modern rig showed the interesting result that the optimum speed made good to windward (Vmg) occurred at an apparent wind angle of between 24 and 25 degrees for a wide range of true wind speeds from 5 to 16 knots. The true wind angle, and hence the tacking angle for the yacht, decreased from 52 degrees in 5 knots of wind to 47 degrees in 16 knots of wind. The VPP indicated that the yacht should reduce sail in wind speeds above 16 knots and other wind tunnel tests were conducted to optimise the reduce sail configurations for the rig.
It is interesting to note that the flat function values given in Table i for the optimum up-wind Vmg sailing condition have values of less than 1.0 at all true wind speeds from S knots and higher. This indicates that the optimum performance from the rig is achieved without sheeting the sails as hard as could be achieved in the wind tunnel.
Off-wind performance
Unlike the prediction of optimum up-wind conditions, there was a considerable variation with wind speed for the apparent wind angles for optimum speed made good down wind, although the optimum true wind angle only varied from 142 degrees in 5 knots of true wind to 155 degrees in 16 knots. The associated apparent wind angles varied from 47 to 106 degrees. These predictions indicate that provided the yacht is sailed at an angle for optimum downwind Vmg speed then the blanketing and interference effects between the sails, discussed earlier, should be avoided. It can, however, be seen that the sails were not tested to the lowest apparent wind angle predicted by the VPP and it may not be possible to set the sails at these angles. Other sail combinations were tested for these conditions
HEEL CORRECTIONS
It can be seen from the photographs of the tests that they were conducted with the model upright. This simplifies the analysis of the data since the sail coefficients input to the VPP are those in the plane normal to the mast, which is the measurement plane for the upright condition. Algorithms within the VPP, based on reference 5, calculate the driving force and heeling moment for the heeled yacht. Essentially there is a reduction in the driving force with heel angle, as shown in reference 2, that may be considered to be related to the reduction in apparent wind angle in the heeled plane from that in the horizontal plane.
Tests have been conducted with a model upright and heeled, with the same sail configuration, and the data were corrected to produce lift and drag coefficients in the plane normal to the mast. The corrections used the same algorithms as those in the VPP with due account taken of the horizontal measurement plane for the driving and heeling forces. In general there was good correlation between the data from the upright and heeled tests, subject to some variability in reproducing the sheeting conditions in both tests.
9 CONCLUSIONS
The tests produced consistent data with clear differences between the classic and modern rigs. The results enabled informed decisions to be made regarding the sail conf gurations for the designs under consideration. Sail coefficients were fitted to the data such that they matched the measured forces and could be input to a 'VPP. This program then produced reliable predictions of the potential performance of the different yacht designs and enabled the relationship between their stability and sail area to be assessed.
ACKNOWLEDGEMENTS
Thanks are due to Gerard Dijkstra and Partners for permission to publish the test data and to staff at the Wolfson Unit MTTA for their assistance with the tests and analysis.
REFERENCES
OLWER, J. C. and CLAUGHTON, A. R., "Development of a Multifunctional Velocity Prediction Program (VPP) for Sailing Yachts", RINA International Conference CADAP 95, Southampton, September 1995.
CAMPBELL, I. M. C., "Optimisation of a sailing rig using wind tunnel data", SNAMIE 13th Chesapeake Sailing Yacht Symposium, Annapolis, January 1997.
CAMPBELL, I. M. C., "The performance of offwind sails obtained from wind tunnel tests", RJNA International Conference on the Modem Yacht, Portsmouth, March 1998.
HAZEN, G. S., "A Model of Sail Aerodynamics for Diverse Rig Types", SNAME New England Sailing Yacht Symposium, New London, Conn., March 1980.
KERWIN, J. E., "A Velocity Prediction Program for Ocean Racing Yachts Revised to February, 1978", H. frying Pratt Ocean Race Handicapping Project, MTT Report No. 78- 11, Cambridge, Mass., March, 1978.
-99-- True wind speed (Vr) - knots 5 6 7 8 9 10 12 14 16 20 25 Optimum up-wind Vmg sailing condition degrees 52.4 51.5 50.6 49.849.048.3 47.5 47.3 47.447.4 48.2 Ba degrees 24.0 24.024.1 24.3 24.5 24.8 25.5 26.3 27.4 29.1 31.6 V knots 5.81 6.69 7.42 8.01 8.48 8.85 9.43 9.8510.17 10.46 10.7 Vmg knots 3.54 4.17 4.71 5.175.565.89 6.37 6.68 6.88 7.09 7.14 heel degrees 6.2 8.3 10.4 12.5 14.4 16.1 18.9 21.3 22.224.4 24.8 reef 1 1 1 1 1 1 1 1 0.95 0.91 0.74 flat 0.85 0.82 0.78 0.74 0.7 0.66 0.58 0.5 0.49 0.41 0.49
Beam reaching with true wind angle B = 90 degrees degrees 29.7 30.932.3 33.935.837.7 41.3 44.847.9 53.1 58.0 V knots 8.66 9.810.7111.36 11.79 12.12 12.59 12.9313.2113.6814.11 heel 8.6 11.6 14.2 16.4 18.1 19.5 21.7 22.622.7 23.3 24.2 reef 1 1 1 1 1 1 1 0.96 0.890.79 0.7 flat i 1 0.97 0.92 0.880.840.76 0.73 0.76 0.8 0.83
Optimum down-wind Vmg sailing condition Pt degrees 142.5142.9143.3143.3144.1145.5149.1152.6154.6155.1168.5 Pa degrees 46.948.4 50.7 53.3 57.2 62.877.7 93.6105.9118.6155.1 V knots 6.81 7.99 9.029.9610.6711.1311.6312.02 12.513.5513.76 Vmg knots 5.4 6.37 7.23 7.98 8.64 9.179.9810.6811.2912.2813.48 heel degrees 3.1 4 4.8 5.7 5.9 5.5 4.2 3.3 2.8 3.3 2.7 reef I i I i 1 1 1 1 1 1 1 flat 1 1 1 1 1 1 1 1 1 1 1
Table 1. Results from VPP calculations with the modern rig
- 10c- Figure 1. Photographs of the classic Schooner rig model in the wind tunnel
Figure 2. Photographs of the modern Schooner rig model in the wind tunnel
tcD-- Classicrig Modern rig 1.4 1.4 I I o z Ba=25deg z ci Ba25deg 1.2 o Ba3Odeg 1.2 O Ba=3Odeg a) :- L. Ba36deg - Ba36deg C,, X Ba=45deg X Ba45deg o Ba6Odeg : °Ba=6Odeg X i0.8 -VPPfit - -VPP fit
C C 0.6 0.6 Co oa) oI- 0.4 C) C . 0.2 L. o o
0.0-i 0.0 5-i.10 15 20 25 30 35 5 10 15 20 25 30 35 Heeling moment Heeling moment at unit wind pressure - kN.m at unit wind pressure - kN.m
Figure 2. Variation of driving forces with heeling moments including windage
Classic rig Modern rig 0.7 0.7 0 Ba=25deg D Ba=25deg a) a) : o Ba=3Odeg o) : o Ba3Odeg D) - Ba36deg 0.6: Ba=36deg 0.6 X : X Ba=45deg : X Ba45deg Ba6Odeg : o Bar6Odeg X : O .Her5lm gO.5-.---He=49m 0.5 .( I Datafit for VPP I Data fit for VPP / o 0.4 X a) a) OQ,3-D 003 C C 0 (1.) 0.2::/ a) o 0.1 0.1 o o 0.0UUNU 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Square of lift coefficient Cl2 Square of lift coefficient Cl2
Figure 3. Variation of drag coefficient with the square of lift coefficient, excluding windage - Classic rig Modern rig 2.0 2.0
1.8 1.8 o 1.6 1.6 P1.
1.4 1.4 o 1.2 1.2_"À.______r ______1.0 1.0 A V o Ba25deg - Ba=25deg ir D oBa=3üdeg O Ba3Odeg 0.8W 0.8 A Ba=36deg I1aA Ba = 36 deg VA X Ba=45deg X Ba=45deg 0.6 o Ba6Odeg - 0.6VLUI Datafit for VPP Datat forVPP 0.4-. 0.4-roI 10 20 30 40 50 60 70 10 20 30 40 50 60 70 Apparent wind angle - degrees Apparent wind angle - degrees
Figure 4. Variation of lift coefficients with apparent wind angle, excluding windage
Modern rig Modern rig 2.5 0.30 I I D (t Cdfrom tests Q) A Clfrom tests Q) 2.0 0.25 o -- - CdFit toVPP windagemodel co (o -- - ClFit to VPPwindagemodel c, 1.5 0.20 C ----n Q) Q) .c:1.0 u____ 0.15 y-- C Driving force / ---- AU , --- W DA Heeling force 0.5 -' 0.10 '1' - - - - VPP cosine function fit C .\JIDsine function fit co
U.0 ii I 0.05 L..
i5 -0.5 0.00 o SIL. Í1 -0.05 0 20 40 60 80 100 120 0 20 40 60 80 100 120 Apparent wind angle - degrees Apparent wind angle - degrees
Figure 5. Variation of model windage forces Figure 6. Variation of windage coefficients with apparent wind angle with apparent wind angle
- Classic rig Modern rig 38 38
36 36 9 o -J D * - 4> A -g)o34 tA O) Lb a5a) - D X a) o a) > 4>A -zo D 4> -t 32 32 - n b 00 X w - b o a) o - (9 9-o 0) C30 a) - C/) z DBa25degQE oa) 28 28 DBa=25deg- 4> Ba = 30 cieg oBa=3odeg - Ba=36deg 26 A Ba = 36 deg - 26 - xBa=45deg o Ba =60 deg oBa6Odeg 24 t t lit.11 tit t t ii ti, ¡t li 24 It,iiiitt humt titi. 0.8 1.0 1.2 1.4 1.6 1.8 2.0 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Heeling force coefficient Cy Heeling force coefficient Cy
Figure 7. Variation of centre of effort height above DWL with apparent wind angle
Classic offwind rig Modern off-wind rig 2.0 2.0 - D Beta6O D Ba75deg
- oBeta9O - 1.8 - 1.8 o Ba=9üdeg - X Betal2O A BalO5deg 0 Beta=135 1.6-/ X Bal2Odeg - A Beta=150 - 1.6 Datafit for VPP : o Bal35deg - - 1.4 1.4 -+--Datafit for VPP
1.2 01.2 z a) \ o 1.0 1.0 a) o 0.8 0.8 -J \\\\\ 0.6 0.6
0.4 0.4
0.2 - 0.2
it., uhu ti i ti 0.0 40 60 80 100 120 140 160 40 60 80 100 120 140 160 Apparent wind angle - degrees Apparent wind angle - degrees
Figure 8. Variation of lift coefficient with apparent wind angle, excluding windage
_kDL- Classic offwind rig Modern off-wind rig 1.0 1.0 B - D 0.9 0.9 NA 0.8A 0.8 0.7/ 0.7 o -o 0 0.6 0.6 c a) o 0.5 o C) C)0.4 ot- o 0.3 0.3 D Ba6Odeg D Ba=75deg Ba9Odeg - 0.2 0.2 Ba=9üdeg X Bal2Odeg A Ba=lO5deg
oBal35deg < 0.1 0.1 A Bal5Odeg O Bal35deg Datafit for VPP --4--- Datafit for VPP 0.0 IÌIÌti__ 0.0 40 60 80 100 120140160 40 60 80 100 120 140 160 Apparent wind angle - degrees Apparent wind angle (degrees)
Figure 9. Variation of drag coefficient with apparent wind angle, excluding windage
14
12
10
-4-- Up-wind condition B Reaching, Bt = 90 deg a-- Down-wind condition
0 4 8 12 16 20 24 True wind speed Vt - knots Figure 10. Variation of predicted speedic with true wind speed, from VPP calculations
"The Influence of the Bowshapeon the Performance of a Sailing Yacht"
by
Dr ir J A Keuning * ROnnink* A Damman *
Abstract
In this paper some results of two studies carried out at the Shiphydromechanics Department of the DeIft University of Technology: one on the influence of an increase of stem steepness of a sailing yacht and another, which was largely carried out by T.J.E.Tincelin as part of his master thesis at Deift University of Technology, on the effect of above waterline bowflare are presented.
To investigate the influence of bow steepness a model of the Deift Systematic Yacht Hull Series (DSYTiS) has been used as a parent model of a new small subseries with two additional derivatives each with increased bow steepness. The influence hereof on both the calm water resistance and the added resistance in head waves has been investigated. To investigate the influence of bow flair two models of a typical "Open 60" designs have been used: one "normal" and one with almost "no-flare" in the bow sections. These have been tested in calm water and in both head- and following-waves to investigate the effects of this difference in bow shape on the calm water resistance, the added resistance inwaves and also on the relative motions at the bow. The results will be presented and some comparisons with calculations made. Also some general conclusions with respect to resistance, performance and safety will be drawn
List of Symbols.
A. wave length [m] Raw = added resistance in waves [NI R'aw = Raw/Ç2 [N/rn2] Rt total resistance = Rf+ Rr [N] Rf = frictional resistance [N] Rr = residual resistance [NJ
Fn = Froude number = y / si(g*Lwl) E-]
w = wave frequency _1uJ = wave amplitude [m]
* Shiphydromechanics Department of the Delfi University ofTechnology i - Introduction.
Stimulated by all kinds of reasons, ranging from the assumed influences of certain rating rules to design considerations driven sometimes by facts or just by fashion, there is a clear trend to be observed over the past decade to steeper and steeper stems in particular with the racing- and other high-performance yachts. The influence ofthis on a particular design may be found in an increased waterline-length with a constrained length-over-all, which on its turn, may lead, assuming constant displacement, to an increased fineness of the waterlines in the forepart ofthe yacht, i.e. a decrease in the angle which the waterlines make with the centerline of the hull (waterline entrance angle). Under the same assumption of constant displacement and overall length, there is also an effect on the prismatic coefficient (Cp decreases), the relative longitudinal position ofthe center ofbuoyancy (LCB moving aft), the centroid of the waterplane area (LCF also moving aft) and pitch radius of gyration. The possible influence of all these changes on the performance of sailing yachts when compared in a more systematic way appeared to be not readily available in the open literature. Also this bow steepness had not been a parameter under consideration in the Deift Systematic Yacht Hull Series (DSYHS) because all the design variations investigated within the DSYEIS so far were obtained by an afme-transformation technique,which implies that the bow steepness for all models originating from one parent model is more or less constant.Therefor it was decided to extend the DSYHS with two additional models, each with increasing bow steepness in comparison with their parent model, which was the parent of Subseries 4, i.e. the IMS-40. Because part from the reasoning behind the possible benefits of the steeper stem originated from a supposed reduction in the added resistance of these ships in waves it was decided not only to test these models for their calm water resistance but to include tests in regular head waves to investigate their differences in that respect also. Another aspect of the possible influence of the bowshape on the performance was investigated using two models not derived from a parent of the DSYHS, i.e. two model variations of a contemporary Open 60 design. This study was aimed at what the influence would be on resistance both in calm water and in waves of the amount of flare in the bowsections above the design waterline. In this case an attempt was made to keep the shape of the foreship below the waterline as identical to each other as possible and make a rather drastical change in the flare of the section above it. The philosophy behind reducing the flare of the bow sections is that here also the added resistance due to sailing in waves is reduced because sailing in (head) waves the amount of energy dissipated is reduced because less damping waves areradiated from these less volumous and beamy bow sections when these are performing large relative motions with respect to the disturbed water surface. However these less volumous bow sections will also imply a lower (nonlinear) pitch restoring moment, which in its turn could lead to higher relative motions at the bow. If this is the case this then could be particularity hazardous when sailing in following waves conditions where broaching (or even pitch poling) could occur due to (increased) deck submergence (bow diving). Therefor in that particular research project also tests in waves have been performed and well both in head- and following waves. In these tests the added resistance has been measured and also theheave-, pitch- and relative motions at the bow. So information is now presented on the influence of (increasing) stem steepness and bow fmeness below the design waterline whilst keeping the deck profile more or less constant and on the tiare of the bow sections keeping the underwater part andbow profile more or less constant. Albeit on different boats. 2 - The Experiment
2 - i The Models.
As explained above two different sets of models have been used for the two different parts of this project:
The first set is composed of a parent model and two systematic variations thereof. The parent model of this set is the parent mode! of Subseries 4 of the DSYHS (model # 44) and is known as the IMS-40. The two additional models are designed with increasing stem steepness in two equal steps. The changes in the hull lines, originating from the changes in the stem steepness, have been restricted to the front half of the models only, i.e. from stern (ordinate O) to midship (ordinate 5) all three models are exactly identical. This choice implied that the waterline length of the three models increases significantly with stem steepness, because the overall length has been kept constant, and this again results in small changes in volume of displacement, Cp, etc. The two newly derived models are introduced into the DSYHS as models # 51 and # 52. The lines of these three models are presented in Figure i.
#44
#51 U1!i1HiiU1
#52
AAA A
Figure 1 Linesplans of the bow steepness variations within the DSY}IS, i.e. the models # 44, # 51 and # 52
.- tog- From these lines it becomes immediately obvious how the forebody shape between these three models changes when the stem is steepened, in particular it isleading to hollow waterlines with the steepest stem (model # 52) and since the deck profile ismaintained also some flared sections forward. The main particulars of these three DSYHSmodels are presented in the Table i below:
TABLEi
Model number DSYHS # 44 #51 #52
Length over all (m) 12.31 12.50 12.43 Length on the waterline (rn) 9.98 10.27 10.53 Beam on the waterline (in) 3.02 3.02 3.02 Canoe body draft (m) 0.68 0.68 0.68 Volume of Displacement (m3) 8.08 8.21 8.28 Waterp lane area (m2) 20.3 20.5 20.6 Wetted Area canoe body (m2) 23.8 24.3 24.5 LCB (1/2 Lord) (in) -0.32 -0.25 -0.21 LCF (1/2 Lord) (in) -0.63 -0.58 -0.50 Prismatic Coefficient (-) 0.558 0.551 0.543 Pitch Radius of gyration (%Loa) 20 20 20
From the data presented in this Table I the differences between the three models, originating from the choices made with regard to the transformation process, are obvious.
The second set consists of two models used for the investigation on the effect of bow fiare and are derived with the aid of Groupe Finot from Paris (France) along the lines of an Open 60. The basis design is with the "usual" flare in the bow sections and the second design is with no flare in the bow sections. Although this may not be a totally realistic design it was used for making the possible effects more significant. Great emphasis was placed on keeping the underwater part as identical as possible without creating an unrealistic design. The linesplans of the two models are presented in Figure 2. The main particulars of the two models are presented in Table 2.
TABLE 2
Model# 60-1 60-2
Length over all (m) 18.28 18.28 Length on the waterline (m) 1ó.97 16.97 Beam on the waterline (m) 3.89 3.91 Canoe body draft (m) 0.41 0.41 Volume of Displacement (mi) 11.02 10.95 Waterplane Area (m2) 45.0 ,IA Wetted Area (m2) 47.1 46.5 Long. .Pos.Centre Bouyancy (m) -0.17 -0.18 Long.Pos. Centr.Waterplane Area(m) -0.75 -0.82 Prismatic Coefficient Cp (-) 0.57 0.57 Pitch Radius of gyration (%Loa) 22 22 -Ito- As may be seen from this Table 2 the two models are almost identical as far as the hydrostatic parameters of their canoe bodies are concerned. The difference between the two models are found in the above water part of the hulls as becomes apparent from the linesplans and the bodyplans of the two models as these are presented in Figure 2.
60-1
60-2
10 8 6 4 2 3 5 79
Figure 2 The linesplans and body plans of the bowflare models
2-2The Measurement Setup and Scheme
- ita . pet n+hh-.+ . ockr1 Laboratory. The dimensions of this tank are: Length 145 meters, width 4.5 meters and maximum attainable waterdepth 2.5 meters. The towing carriage is capable of reaching speeds up to 8 rn/sec. At one side the tank is equipped with a hydraulically activated wavemaker capable of generating regular and irregular waves. During the tests all models were connected to the towing carriage in such a way that they were free in pitch and heave but restrained in all other modes of motion. This was established using the so-called "nutcracker" device customary for seakeeping tests, which connects the model to the carriage solely through the exact position of the Center of Gravity of the model. For the DSYHS this is not the way in which the calm water resistance is measured normally.
s ( - Therefor model # 44 has been remeasured in the scope of the present study to make sure that all models have been measured using exactly the same procedure for each of them. The standard procedure of the Delfi Shiphydromechanics Laboratories with regard to turbulence stimulation has been used in these experiments also, i.e. the performing the measurements with both three half- and three full width carborundum stripes on the forepart of the model. The extra resistance caused by these stripes has been determined by taking the difference between the measured resistance using the half and full width and subtract this difference twice from the measured model resistance with the full width stripes to yield the desired measured model resistance. During the tests in waves the waves were measured using three double strings type waveheight measurement devices and the motions of the models were measured using an optical six degrees of freedom tracking device. For the tests in following waves the models were lifted out of the water to make sure that the generated waves could pass underneath the model prior to the measurement without being disturbed by the presence of the model.
The measurement scheme used for the three DSYHS models was:
1 calm water resistance in the Froude range from Fn = 0.15 to n = 0.60 with half and full width of the carbo rundum strips. 2Added resistance measurements at two different forward speeds corresponding to Fn 0.265 and Fn = 0.3 25 with regular head waves ranging in length from X = 0.7 Lwl to X 3.5 Lwl and with wave amplitudes yielding two different wave steepness values i.e. X / 2Ç = 30 and = 40 respectively.
The measurement scheme for the two Open 60 models was:
i calm water resistance tests in the Froude range from Fn0.15 to Fn0.75 with both half and full width of the carborundum strips 2Motion and added resistance measurements in regular head waves at Fn = 0.35 (typical upwind condition) with wave ranging from X0.7* Lwl to X3.0* Lwl and a wave steepness between X / 2= 20 and X / 2= 50 3 Motions and added resistance measurements in regular following waves at Fn = 0.65 and Fn = 0.75 (typical reaching conditions) with waves ranging from X = 1.2* Lwl and X = 2.1 * Lwl and wave steepness between X / 2= 20 and X / 2Ç1 = 40
The measurements with the Open 60 models have been carried out with and without correctional trimming moments for the sail (driving) forces and at both O and 20 degrees of heel (no leeway). For the extrapolations of the measurement data to full scale Froude's extrapolation method has been used together with the 1TTC-57 formulation for the extrapolation coefflcient Cf. No formiactor has been applied with the DSYHS models because they turned out to be below 0.04 and not using them is consistent within the procedure adopted in DSYHS.
-t2- 3 - Discussion of the Results
3- iCalm water results
In Figure 3 and Figure 4 the results of the extrapolation to full scale of the calm water residual upright resistance of the three DSYHS models are presented. The difference between the two plots originates from a different way of comparing these three models: i.e. Figure 3 is plotted on basis of identical forward speeds in rn/sec for all models, taking the waterline-length ofthe model # 44 as the benchmark to determine the Froude number, so no correction is applied for their difference in actual waterline-length and in Figure 4 the same results are presented but now on a basis of the corrected Froude number based on the actualwaterline-length of each model.
30000
25000
20000
z .- 44 .--51 15000 I. X
10000
5000
0.000 0100 0.200 0300 0400 0 500 0800 Fn (-] (based on model #44) Figure 3 Residuary resistance of the DSYFIS models for identical speeds
30000
25000
20000
#52[
0200 0300 0400 0500 Fn [-J (actual) Figure 4 Residuary resistance of the DSYHS models for actual Froude numbers The differences in Figure 3 are rather obvious: the longer waterline yacht performs better for a given speed. So increasing waterline-length within the constraint of a constant length over all makes sense. Compared on basis of their actual waterline-length and correct Froude numbers however the differences in the residuary resistance between these three design variations become very small. However in the speed range from Fn = 0.35 to Fn = 0.40 (critical speed) and then again above Fn0.50 there appears to be an increasing small benefit for the models with the finer sections forward.
In Figure 4 also the results of an approximation for the residuary resistance are presented using the polynomial expression based on the overall results of the DSYHS as it is presented in Figure 5, Figure 6 and Figure 7.
30000
25000
20000 z ---#44 full scale 15000 # 44 polynorn ial a:
10000
5000
o 0000 0100 0200 0300 0400 0500 0600 Fn (-] (actuaì) Figure 5 Comparison with DSYS model #44 and polynomial nov. 1998.
30000
25000
20000 z --# 51full scale 15000 # 51polynOEnlal fra:
10000
5000
o o000 0 100 0 200 0300 0 400 0 500 0600 En [-] (actual) Figure 6 Comparison with DSYS model #51 and polynomial nov. 1998.
-M Although the results are in general in quite close agreement with the measured data it is interesting to note that these expressions yield a reverse trend for the three models in the speed range around Fn 0.40. It should be noted that in this polynomial expressions no explicit parameter like the waterline entrance angle is present and the differences in forebody shape are only taken into account using an implicit approach through LIB and Cp etc.
30000
250(X)
20000
z if 52 full soele 15000 if 52 polynomial
10000
5000
o 0000 0.100 0.200 0300 0.400 0.500 0.600 Fn (.] (actual) Figure 7 Comparison with DSYS model #52 and polynomial nov. 1998.
The extrapolated results for the upright total resistance of the two Open 60 designs are presented in Figure 8 for the condition during the measurements in which no correction moment was applied for the trimming moment excerted by the driving force generated by the sails. In this extrapolation to full scale a form factor of around k = 0.11 (and slightly higher for model 60-1) has been used. This value for the form factor was derived from the measurements by using a so-called Prohaska plot.
16000
14000
12000
10000 z 4--model 60-1 6000 model 60-2
6000
4000
2000
o o 01 0.2 0.3 04 0.5 0.6 0.7 08 Fn (-] Figure 8 Upright resistance of model 60-1 and 60-2 without trim moment -t t'- As may be seen from this Figure 8 the differences between the two designs are very small although model60-2appears to have in general a reduction on total resistance of around1.2% compared to60-l.This is to be expected because of the close similarity of the models up to the design waterline and the small difference will be induced by the slightly lower displacement and wetted area of the model60-2.
The difference in the above water shape between these two models becomes only significant when the results of the resistance measured with the longitudinal trimming moment is applied during the measurement, i.e. the moment that is excerted on the boat due to the driving force (equal to the resistance at that speed) in the center of effort of the sails. These results are presented in Fi.gure 9
16000 -
14000
12000
z 4--model60-1 - I-8000 model60-2
6000
4000
2000
o 0 0.1 0.2 03 0.4 0.5 0.6 0.7 0.8 Fn [-] Figure 9 Total resistance of models60-1and60-2with trimming moment applied
From these results it can be seen that the model60-2has a definite advantage over the60-1in particular when this trimming moment becomes substantial, i.e. at speeds above Fn = 0.30. In the entire speed range of 10 to 15 knots model60-2,with the "low volume bow", has circa 8% less resistance. This difference in resistance is caused by the differences in sinkage and running trim: the sinkage of model60-2is significantly less and the running trim, which can be quite substantial for these kinds of yachts (i.e.1.5 - 2.0 degr.), is lower at the higher speeds. At the highest speeds the bows are clear of the water and the differences diminishes. These results will undoubtedly be influenced by the designs under consideration, i.e. the Open 60.The trends observed on the resistance of these models however will probably also be valid, albeit to a smaller extend, on other designs. 3- 2Head waves added resistance
Some results of the added resistance measurements in head waves with the three DSYHS models are presented in the Figure 10 and Figure Ii as being typical for all other results also. In these figures only the results of R'aw for one forward speed are presented at two different values for the wave steepness, i.e./2Ç = 30 and X/2c = 40.
0,3
0.25 - -A.
0.2-
44 0.15 --A- #51 (5
0.1 -
0.05
o 05 1.5 2 25 3
XJL (-] Figure 10 Added resistance in head waves at Fn = 0.325 and ?/2= 30
0.3
0.25
0.2
.-- # 44 015 - -U - # 51 (5
0.1
0.05
o 05 1 1.5 2 2.5 3 AiL (-J Figure 11 Added resistance in head waves at Fn = 0.325 and ?/2= 40
-t t-- These results are in contradiction maybe to what was expected; i.e. the model with the finest bow (i.e. # 52) has the lowest added resistance in waves. The difference between the three models appears to be small however, in particular when an accuracy band around the measured values is considered, then no significant difference may be found. Thereis inevitably some inaccuracy in these Raw measurements because this "added resistance in waves" is obtained by subtracting two rather large quantities: i.e. the calm water resistance at the forward speed under consideration and the averaged time integrated (higher) resistance in waves at the same speed. This later quantity on its turn is determined byintegrating and averaging a rather oscillatory force signal over a large number of wave encounter periods. So the trend between the added resistance differences between the models is not very consistent. What may be observed from the results also, and this has been found before by other authors such as Hirayama for an IACC yacht in head waves (Ref [ 4]), is that there appears to be a dependency of the added resistance RAU on the wave steepness in such a way that the added resistance in waves RAU, i.e. R'aw = Raw I Ç, decreases with an increasing wave steepness, i.e. increasing Ç for a given ?. This nonlinearity in Raw, which may be introduced by the changes in displaced volume of the bowsections which is brought into contact with the water whilst the bow is performing large relative motions. In this respect the three DSYHS models have a contradictionary trend with respect to their bow fmeness, i.e. the underwater part (and so the waterlines) are "stretched" with increasing bow steepness but in the sametime the flare of the sections above the waterline is increased, because the deck profile has been kept more or less constant. This may be seen from the body plans of these models as presented. Alsothe tests for all three models have been performed on exactly the same model speed, which actually implies a somewhat lower Froude number for the longer models. This phenomenon is maybe even more clearly demonstrated by the results obtained from the tests with the two Open 60 models in head waves as presented in Figure 12, because between these models only the flare of the bow sections has been changed.
40
35
30 -
25 .--4%steep. model 60-1 .---2% steep model 60-1 a -4%steep model -s.»2%steep. Model6O-21
3 35 4 4.5 5 55 6 Wave frequency [radis] Figure 12 Added resistance in head waves at Fn0.35 of model 60-1 and 60-2
( tes- From Figure 12 it is obvious that almost over the entire frequency range investigated the added resistance in waves of model 60-1 is higher than the added resistance of model 60-2. Although the RAO's are (again) strongly reduced with increasing wave steepness the difference between these two models increases also with increasing wave steepness and the maximum difference in the RAO may become as large as 20% for certain wave lengths. The effect on the added resistance of a larger increase in the momentary submerged (sectional) volume at the bow of model 60-1 while performing large relative motions apparently overrules the effect of the larger relative motions at the bow of model 60-2 because these were found to be significantly larger for model 60-2 when compared to model 60-l. This was confirmed by analysis of the time history of the resistance/surge force signal from the transducers. The larger relative motion for 60-2 was conform expectation. It should be noted though that no deck submergence occurred during these tests with either of the two models. The amount of water on deck however was considerably larger for model 60-1 than for model 60-2 possibly due to significant difference in the dynamic swell-up and spray generated.
3 - 3 Following waves
The added resistance in following waves has been measured with and without the correctional trimming moment to simulate the longitudinal moment caused by the driving forces on the sails. Without this correction the differences in added resistance are generally small, although again increasing with the wave steepness similar to the head waves condition. With the longitudinal correction moment applied the differences between the two models become more significant. This is probably largely due to the differences in resistance between the two models when the bow is being trimmed down, i.e. the model with bow flare (60-l) suffers more than the model without (60-2) Results are presented in Figure 13.
4000
3500
3000 z ø 2500 L. '--model 60-1 3% L) L C ---model 60-1 4% L a--model 60-1 5% 2000 - - - model 60-2 3% - - model 60-24% - - *- - model 60-2 5% 1500
1000
500
36 3.7 3.8 39 4 4.1 4.2 4.3 4.4 Wave frequency (radis] Figure 13 Added resistance in following waves at Fn = 0.65 model 60-1 and 60-2 Although these results are not presented here the tests in following waves at the highest speed, i.e. Fn = 0.75 showed even larger differences between the two models. As far as the safety of the yacht is concerned the results of the vertical displacement of the bow for the two models in following waves is presented in Figure 14. From this Figure it is obvious that the bow of model 60-2 at higher speeds is almost 10 cm deeper in the water than the flared bow of model 60-1. Although this may seem a small difference in steep following waves this made a tremendous difference in the likelihood of bow diving and green water on deck.
25
20
E o 15 o -4--model 60-1 10 ---I- model 60-2 o o, 5
CI)
-10 Fn [-] Figure 14 Bow sinkage in following waves with trim moment applied.
In the speed range from 8 to 17 knots the model with the thinner bow (model 60-2) performed some 15 to 20% larger relative motions at the bow compared to the flared bow (model 60-l) in particular for the longer and the steeper waves. 3 -4 Conclusions
As a conclusion of both studies reported here it could be stated that increasing the stem steepness implying stretching the waterlines in the underwater part of the hull while keeping the above water part of the bow more or less constant does not seem to have a large influence on the calm water resistance as long as the lengthening of the waterline is being taken into account properly. The fine waterline entry does not influence the added resistance due to waves to a large extend because the relative large relative motions at the bow make the shape of the bow above the still waterline at least as important as the shape below it. Reducing the flare in the bow sections creates a significantly larger difference in added resistance, both in head and in following waves. But with respect to the safety of the yacht the increase of the relative motions at the bow may cause an increased probability of deck submergence.
References
Keuning J A, Sonnenberg U.B., Approximation of the Hydrodynamic Forces on a Sailing Yacht based on the Delfi Systematic Yacht Hull Series Report 1 175-P Delft Shiphydromechanics Laboratory and 1 5-th International FHSWA Symposium on Yacht Design and Construction (Amsterdam) November 1998
Tincelin T.J.E., Influence of the Hull Form above the Static Waterline on the Seakeeping Properties o f Open 60's Master Thesis, Shiphydromechanics Department Deift University of Technology June 1999
Levadou M M D, Added Resistance in Waves of Sailing Yachts Master Thesis, Shiphydromechanics Department Deift University of Technology July 1995
Falsone J Marine tests of the PACT base boat in following waves The 13-th Chesapeake Sailing Yacht Symposium SNAME 1997 GERARD DIJKSTRA & PARTNERS LUTRA DESIGN GROUP NAVAL ARCHITECTS & MARINE ENGINEERS
YACHTS-VAN LENT YACHTS
STAND FAST YACHTS
NIEUWBOUW & ONDERHOUD
ROYAL 1-IUTSMAN .*SHTPYARD